CN101277368B - Image processing apparatus, image forming apparatus, image processing system, and image processing method - Google Patents

Image processing apparatus, image forming apparatus, image processing system, and image processing method Download PDF

Info

Publication number
CN101277368B
CN101277368B CN2008100900071A CN200810090007A CN101277368B CN 101277368 B CN101277368 B CN 101277368B CN 2008100900071 A CN2008100900071 A CN 2008100900071A CN 200810090007 A CN200810090007 A CN 200810090007A CN 101277368 B CN101277368 B CN 101277368B
Authority
CN
China
Prior art keywords
registered images
similar degree
image processing
image data
unit
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN2008100900071A
Other languages
Chinese (zh)
Other versions
CN101277368A (en
Inventor
大平雅和
吉村秀义
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sharp Corp
Original Assignee
Sharp Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Sharp Corp filed Critical Sharp Corp
Publication of CN101277368A publication Critical patent/CN101277368A/en
Application granted granted Critical
Publication of CN101277368B publication Critical patent/CN101277368B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Control Or Security For Electrophotography (AREA)
  • Image Processing (AREA)
  • Facsimiles In General (AREA)
  • Facsimile Image Signal Circuits (AREA)
  • Image Analysis (AREA)
  • Accessory Devices And Overall Control Thereof (AREA)
  • Processing Or Creating Images (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

A storage process section stores, into an exceptional reference image list, discrimination information for discriminating a reference image which permits image processing and a reference image which forbids or restricts image processing. In case where a maximum value of a similarity calculated by a voting process section is not less than a threshold value, a control section determines whether a reference image having a maximum similarity is the reference image which permits image processing or a reference image which forbids or restricts image processing and determines whether or not it is necessary to forbid or restrict image processing with respect to input image data in accordance with a result of the determination. This makes it possible to prevent erroneous determination from forbidding execution of image processing based on input image data or to prevent erroneous determination from allowing execution of image processing based on input image data.

Description

Image processing apparatus, system, method and image processing system
Technical field
The present invention relates to judge the similar degree of the registration image data of input image data and registered in advance, the image processing apparatus, image processing system, image processing system and the image processing method that carry out the control corresponding with judged result.
Background technology
In the past, use will be compared by the registered images that scanner reads input image data that original image obtains and registration in advance and judge both similar degree, according to the technology of judged result control for the processing (for example, duplicating, transmission, editor etc.) of input image data.
The luminous letter reader) etc. as the determination methods of similar degree, for example known: by OCR (Optical Character Reader: extract password (keyword) from image, thereby carry out method based on the coupling of password; Object images is defined in the image of bill with ruling, carries out method (with reference to patent documentation 1 (Japanese publication communique spy opens flat 8-255236 communique (on October 1st, 1996 is open))) based on the coupling of the feature of ruling; And according to the color sensation of input image data and the registered images method (with reference to patent documentation 2 (Japanese publication communique spy opens flat 5-110815 communique (on April 30th, 1993 is open))) judged etc. that distributes.And, following technology is disclosed in patent documentation 3 (Japanese publication communique spy opens flat 7-282088 communique (October 27 nineteen ninety-five is open)), promptly, generation is selecteed descriptor from the feature of input document, so that its distortion that produces with respect to digitlization owing to document, or the difference between the document of adjusting in input document and the document database does not change, use this descriptor, with record descriptor and indicated the descriptor database of the lists of documents that comprises the feature of extracting descriptor, import the coupling of the document in document and the document database.In this technology, when descriptor database is scanned, added up to for the voting results of each document in the document database, the document that document that number of votes obtained is the highest or number of votes obtained have surpassed certain threshold value becomes the adjustment document.
And, following technology is disclosed in patent documentation 4 (international open No. 2006/092957 brochure of communique (on September 8th, 2006 is open)), promptly, from digital picture, extract a plurality of characteristic points, set for the local characteristic point of each characteristic point decision of extracting, and from each set of decision, select the part set of characteristic point, with the each several part set selected amount as attendant features, a plurality of combinations according to the characteristic point in the part set, ask invariant respectively for geometric conversion, each invariant of obtaining is made up and the calculated characteristics amount, according to the characteristic quantity that calculates to the document in the database, image is voted, thus the retrieval document corresponding with above-mentioned data image, image.
And, as the technology of controlling according to the judged result of similar degree for the processing of input image data, for example known following technology, promptly, form the bank note that device carries out or the forgery of marketable securities in order to tackle coloured image, according to from the detected pattern of input image data, judge whether input image data is the view data of (registered images) such as bank note or marketable securities, under the situation of view data that is registered images, to the additional specific pattern of the image of output, so that can determine to carry out the image processing system that duplicates from the image of output, perhaps be coated with bad copy image, perhaps forbid duplicating action.
Yet, judging that whether input image data is in the image processing apparatus of view data of registered images, because the restriction of image reading accuracy etc.,, the condition when reading breaks because causing erroneous judgement.
Therefore, in above-mentioned technology in the past, judging under the situation of view data that input image data is a registered images, adding specific pattern or be coated with bad copy image and limit contents processing, perhaps forbidding duplicating the processing of action.Therefore, although, be judged as the input image data of the view data of registered images by mistake, can not carry out user's treatment desired for not being the view data of registered images originally.
And, opposite with above-mentioned technology in the past, though consider that judging be under the situation of view data of registered images, (for example allow carries out image processing, duplicate, print, electronics is posted a letter, facsimile transmission, filing, the correction of view data, editor etc.), and under the situation of judging the view data that is not registered images, forbid or limit the structure of carries out image processing, but, in this structure, be judged as under the situation of view data that input image data is a registered images in mistake, although formerly should forbid or limit carries out image processing, allowed the execution of image processing.Therefore, the danger of the problem on the safety such as leakage that produce confidential information for example arranged.
Summary of the invention
The present invention puts in view of the above-mentioned problems and finishes, its purpose is, in the similarity (similar or not similar) of judging input image data and registered images, and according to judged result, whether decision will be forbidden or limit in the image processing apparatus of execution for the image processing of input image data, prevent following unfavorable condition, promptly, because erroneous judgement is disconnected, the former image processing for input image data that should be allowed to carry out can not be performed, perhaps because erroneous judgement is disconnected, the former processing that should be under an embargo or limit the input image data of execution is performed.
In order to solve above-mentioned problem, image processing apparatus of the present invention comprises: the input data obtain the unit, obtain input image data; The Characteristic Extraction unit extracts the characteristic quantity of described input image data; Stored registered images characteristic quantity memory cell with obtain from the registered images of being obtained the characteristic quantity of registered images by the external device (ED) that can be connected communicatedly the unit at least one; The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of registered images, thereby calculates the similar degree of two images; And control unit, whether be more than the predetermined threshold value according to described similar degree, whether decision will forbid or limit the image processing of carrying out for described input image data.In described registered images, permission registered images as the registered images that allows carries out image processing is arranged, with as forbidding or limit the restriction registered images of the registered images of carries out image processing, described image processing apparatus also has identifying information and obtains the unit, this identifying information is obtained the unit and is obtained the identifying information that is used to discern described permission registered images and described restriction registered images from described memory cell or described external device (ED), described control unit is under the maximum of the similar degree that described similar degree computing unit is calculated is situation more than the described predetermined threshold value, at similar degree is that peaked registered images is under the situation of restriction registered images, forbid or limit the image processing of carrying out for described input image data, at similar degree is that peaked registered images is to allow registered images, for the similar degree of restriction registered images under the situation that does not reach the 2nd threshold value bigger than described threshold value, allow to carry out image processing for described input image data, at similar degree is that peaked registered images is to allow registered images, (i) be under the situation more than described the 2nd threshold value at similar degree for the restriction registered images, perhaps, be more than described the 2nd threshold value (ii) at similar degree for the restriction registered images, and described maximum and be under the situation below the setting for the difference between the similar degree of described restriction registered images is forbidden or is limited the image processing of carrying out for described input image data.
Another image processing apparatus of the present invention comprises: the input data obtain the unit, obtain input image data; The Characteristic Extraction unit extracts the characteristic quantity of described input image data; Stored forbid or limit carries out image processing the restriction registered images characteristic quantity memory cell with obtain from the registered images of the characteristic quantity of being obtained the restriction registered images by the external device (ED) that can be connected communicatedly the unit at least one; The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of restriction registered images, thereby calculates the similar degree of two images; And control unit, whether be more than the predetermined threshold value according to described similar degree, determine whether to forbid or limit the image processing of carrying out for described input image data.Described image processing apparatus also has the location registration process unit, this location registration process unit makes the characteristic quantity of the permission registered images that allows carries out image processing, be stored in the 2nd memory cell that described memory cell or described external device (ED) have with the identifying information that is used for discerning described restriction registered images and described permission registered images, described similar degree computing unit calculates the similar degree of described input image data for described restriction registered images, with the similar degree of described input image data for described permission registered images, described control unit is under the maximum of the similar degree that described similar degree computing unit is calculated is situation more than the described predetermined threshold value, at similar degree is that peaked registered images is under the situation of restriction registered images, forbid or limit the image processing of carrying out for described input image data, at similar degree is that peaked registered images is to allow registered images, for the similar degree of restriction registered images under the situation that does not reach the 2nd threshold value bigger than described threshold value, allow to carry out image processing for described input image data, at similar degree is that peaked registered images is to allow registered images, (i) be under the situation more than described the 2nd threshold value at similar degree for the restriction registered images, perhaps, be more than described the 2nd threshold value (ii) at similar degree for the restriction registered images, and described maximum and be under the situation below the setting for the difference between the similar degree of described restriction registered images is forbidden or is limited the image processing of carrying out for described input image data.
Another image processing apparatus of the present invention comprises: the input data obtain the unit, obtain input image data; The Characteristic Extraction unit extracts the characteristic quantity of described input image data; Memory cell and the registered images of obtaining the characteristic quantity that allows registered images from the external device (ED) that can be connected communicatedly of having stored the characteristic quantity of the permission registered images that allows carries out image processing obtained at least one the unit; The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of permission registered images, thereby calculates the similar degree of two images; And control unit, whether be more than the predetermined threshold value according to described similar degree, whether decision allows to carry out the image processing for described input image data.Described image processing apparatus also has the location registration process unit, this location registration process unit makes the characteristic quantity of forbidding or limiting the restriction registered images of carries out image processing, be stored in the 2nd memory cell that described memory cell or described external device (ED) have with the identifying information that is used for discerning described permission registered images and described restriction registered images, described similar degree computing unit calculates the similar degree of described input image data for described permission registered images, with the similar degree of described input image data for described restriction registered images, described control unit is under the maximum of the similar degree that described similar degree computing unit is calculated is situation more than the described predetermined threshold value, at similar degree is that peaked registered images is under the situation of restriction registered images, forbid or limit the image processing of carrying out for described input image data, at similar degree is that peaked registered images is to allow registered images, for the similar degree of restriction registered images under the situation that does not reach the 2nd threshold value bigger than described threshold value, allow to carry out image processing for described input image data, at similar degree is that peaked registered images is to allow registered images, (i) be under the situation more than described the 2nd threshold value at similar degree for the restriction registered images, perhaps, be more than described the 2nd threshold value (ii) at similar degree for the restriction registered images, and described maximum and be under the situation below the setting for the difference between the similar degree of described restriction registered images is forbidden or is limited the image processing of carrying out for described input image data.
Image processing method of the present invention comprises: the input data obtain operation, obtain input image data; The Characteristic Extraction operation, the characteristic quantity of extraction input image data; The log-on data that obtains the characteristic quantity of registered images obtains operation; The similar degree calculation process compares the characteristic quantity of input image data and the characteristic quantity of registered images, thereby calculates the similar degree of two images; And the treatment limits operation, whether be more than the predetermined threshold value according to described similar degree, whether decision will forbid or limit the image processing of carrying out for described input image data.In described registered images, permission registered images as the registered images that allows carries out image processing is arranged, with as forbidding or limit the restriction registered images of the registered images of carries out image processing, described image processing method also has identifying information and obtains operation, the operation that obtains this identifying information obtains and is used to discern the identifying information that allows registered images and restriction registered images, in described treatment limits operation, under the maximum of the similar degree that described similar degree calculation process is calculated is situation more than the described predetermined threshold value, at similar degree is that peaked registered images is to allow under the situation of registered images, forbid or limit the image processing of carrying out for described input image data, at similar degree is that peaked registered images is to allow registered images, for the similar degree of restriction registered images under the situation that does not reach the 2nd threshold value bigger than described threshold value, allow to carry out image processing for described input image data, at similar degree is that peaked registered images is to allow registered images, (i) be under the situation more than described the 2nd threshold value at similar degree for the restriction registered images, perhaps, be more than described the 2nd threshold value (ii) at similar degree for the restriction registered images, and described maximum and be under the situation below the setting for the difference between the similar degree of described restriction registered images is forbidden or is limited the image processing of carrying out for described input image data.
Image processing system of the present invention comprises: image processing apparatus; And for the server unit that this image processing apparatus can connect communicatedly, disperse to have in described image processing apparatus and described server unit: the input data obtain the unit, obtain input image data; Memory cell has been stored the characteristic quantity of registered images; The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of registered images, thereby calculates the similar degree of two images; And control unit, whether be more than the predetermined threshold value according to described similar degree, whether decision will forbid or limit the image processing of carrying out for described input image data.In described registered images, permission registered images as the registered images that allows carries out image processing is arranged, with as forbidding or limit the restriction registered images of the registered images of carries out image processing, described image processing system also has identifying information and obtains the unit, this identifying information is obtained the unit must be used to discern described permission registered images and described restriction registered images from described cell array area identifying information, described control unit is under the maximum of the similar degree that described similar degree computing unit is calculated is situation more than the described threshold value, at similar degree is that peaked registered images is under the situation of restriction registered images, forbid or limit the image processing of carrying out for described input image data, at similar degree is that peaked registered images is to allow registered images, for the similar degree of restriction registered images under the situation that does not reach the 2nd threshold value bigger than described threshold value, allow to carry out image processing for described input image data, at similar degree is that peaked registered images is to allow registered images, (i) be under the situation more than described the 2nd threshold value at similar degree for the restriction registered images, perhaps, be more than described the 2nd threshold value (ii) at similar degree for the restriction registered images, and described maximum and be under the situation below the setting for the difference between the similar degree of described restriction registered images is forbidden or is limited the image processing of carrying out for described input image data.
And, above-mentioned image processing is not limited to the image processing of narrow senses such as the distortion, conversion, coding of image, for example comprises: duplicating, printing, electronics are posted a letter, correction of facsimile transmission, filing, view data, editor, conversion, coding etc. be for all processing of view data.And, above-mentioned input data obtain the unit and for example both can read original copy by scanner and obtain input image data, also can obtain input image data by utilizing software that the form of necessary item input electronic data is made, also can obtain the input image data that directly makes as electronic data, also can obtain the input image data that sends from the device that can connect communicatedly image processing apparatus.
According to above-mentioned structure, the characteristic quantity of the input image data that extraction input data extracting unit in Characteristic Extraction unit obtains.Then, thus the characteristic quantity of the input image data that the similar degree computing unit extracts the Characteristic Extraction unit compares the similar degree that calculates two images with the characteristic quantity of registered images or the registered images characteristic quantity of obtaining the registered images of obtaining from the external device (ED) that can be connected communicatedly the unit in being stored in memory cell.And, in above-mentioned registered images, have as the permission registered images of the registered images that allows carries out image processing with as the restriction registered images of forbidding or limit the registered images of carries out image processing.Then, identifying information is obtained the unit and is obtained the identifying information that is used to discern permission registered images and restriction registered images from memory cell or external device (ED), control unit is under the maximum of the similar degree that the similar degree computing unit is calculated is situation more than the above-mentioned threshold value, at similar degree is that peaked registered images is to allow under the situation of registered images, allow to carry out image processing for input image data, and at similar degree is that peaked registered images is under the situation of restriction registered images, forbids or limit execution for the image processing of input image data.
Thus, because registration has the registered images of permission and restriction registered images, so, under the maximum of the similar degree that the similar degree computing unit calculates is situation more than the above-mentioned threshold value, can be to allow registered images or restriction registered images according to the registered images of similar degree maximum, carries out image processing will be forbidden or limit to decision whether.Therefore, although for example can prevent that also being broken by erroneous judgement is the input image data of restriction registered images for not being the restriction registered images, it as allowing registered images to register, is broken thereby afterwards this input image data is formed to judge by accident.And, in contrast,, it is registered as the restriction registered images, thereby it is disconnected afterwards this input image data to be formed erroneous judgement although can prevent from also to be judged as the input image data that allows registered images by mistake for not being to allow registered images.
And image processing apparatus of the present invention can be following structure also, promptly comprises: the input data obtain the unit, obtain input image data; The Characteristic Extraction unit extracts the characteristic quantity of above-mentioned input image data; Stored forbid or limit carries out image processing the restriction registered images characteristic quantity memory cell with obtain from the registered images of the characteristic quantity of being obtained the restriction registered images by the external device (ED) that can be connected communicatedly the unit at least one; The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of restriction registered images, thereby calculates the similar degree of two images; And control unit, under above-mentioned similar degree is situation more than the predetermined threshold value, forbid or limit the image processing of carrying out for this input image data, it is characterized in that, above-mentioned image processing apparatus also has the location registration process unit, this location registration process unit makes the characteristic quantity of the permission registered images that allows carries out image processing, be stored in the 2nd memory cell that said memory cells or said external device have with the identifying information that is used for discerning above-mentioned restriction registered images and above-mentioned permission registered images, above-mentioned similar degree computing unit calculates the similar degree of above-mentioned input image data for above-mentioned restriction registered images, with the similar degree of above-mentioned input image data for above-mentioned permission registered images, above-mentioned control unit is more than the above-mentioned threshold value at the similar degree for above-mentioned restriction registered images, and be compared under the big situation of the similar degree of above-mentioned restriction registered images for the similar degree of above-mentioned permission registered images, allow to carry out image processing for above-mentioned input image data.
According to said structure, the Characteristic Extraction unit extracts the characteristic quantity that the input data obtain the obtained input image data in unit.Then, the characteristic quantity of the input image data that similar degree computing unit comparative feature amount extraction unit extracts and the characteristic quantity of the restriction registered images that is stored in memory cell or registered images is obtained the characteristic quantity of the restriction registered images of obtaining from the external device (ED) that can be connected communicatedly the unit, thereby calculate the similar degree of two images, control unit is forbidden or is limited the image processing of carrying out for this input image data under above-mentioned similar degree is situation more than the predetermined threshold value.And, this image processing apparatus also has the location registration process unit, this location registration process unit makes the characteristic quantity of the permission registered images that allows carries out image processing and is used for discerning the restriction registered images and allows the identifying information of registered images to be stored in the 2nd memory cell that memory cell or external device (ED) have, similar degree computing unit calculating input image data for the similar degree of restriction registered images and input image data for the similar degree that allows registered images.Then, control unit is more than the threshold value and is compared to for the similar degree that allows registered images under the big situation of the similar degree of restriction registered images at the similar degree for the restriction registered images, allows to carry out the image processing for input image data.
Thus, for the view data that originally is not the restriction registered images, but the similar degree for the restriction registered images is a input image data more than the threshold value, by with it as allowing the registered images registered in advance, can carry out image processing for this input image data.For example, although can prevent for not being the restriction registered images, also by the disconnected input image data of erroneous judgement for the restriction registered images, it is registered as the permission registered images, thereby it is disconnected afterwards this input image data to be formed erroneous judgement, and, can prevent to be forbidden irrelevantly or limit for the image processing of this input image data.
And image processing apparatus of the present invention also can be used as following structure, promptly comprises: the input data obtain the unit, obtain input image data; The Characteristic Extraction unit extracts the characteristic quantity of above-mentioned input image data; Memory cell and the registered images of obtaining the characteristic quantity that allows registered images from the external device (ED) that can be connected communicatedly of having stored the characteristic quantity of the permission registered images that allows carries out image processing obtained at least one the unit; The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of permission registered images, thereby calculates the similar degree of two images; And control unit, under above-mentioned similar degree is situation more than the predetermined threshold value, allow to carry out image processing for this input image data, it is characterized in that, above-mentioned image processing apparatus also has the location registration process unit, this location registration process unit makes the characteristic quantity of forbidding or limiting the restriction registered images of carries out image processing, be stored in the 2nd memory cell that said memory cells or said external device have with the identifying information that is used for discerning above-mentioned permission registered images and above-mentioned restriction registered images, above-mentioned similar degree computing unit calculates the similar degree of above-mentioned input image data for above-mentioned permission registered images, with the similar degree of above-mentioned input image data for above-mentioned restriction registered images, above-mentioned control unit is more than the above-mentioned threshold value at the similar degree for above-mentioned permission registered images, and be compared under the big situation of the similar degree of above-mentioned permission registered images for the similar degree of above-mentioned restriction registered images, forbid or limit the image processing of carrying out for above-mentioned input image data.
According to said structure, the Characteristic Extraction unit extracts the characteristic quantity that input image data is obtained the input image data of obtaining the unit.Then, the characteristic quantity of the input image data that similar degree computing unit comparative feature amount extraction unit extracts be stored in memory cell in the characteristic quantity of permission registered images or the registered images characteristic quantity of obtaining the permission registered images of obtaining from the external device (ED) that can be connected communicatedly the unit, thereby calculate the similar degree of two images, control unit allows to carry out the image processing for this input image data under similar degree is situation more than the predetermined threshold value.And, image processing apparatus also has the location registration process unit, this location registration process unit makes the characteristic quantity and being used for of forbidding or limiting the restriction registered images of carries out image processing discern the identifying information that allows registered images and restriction registered images and is stored in the 2nd memory cell that memory cell or external device (ED) have, and similar degree computing unit calculating input image data are for the similar degree that allows registered images and the input image data similar degree for the restriction registered images.Then, even control unit for the similar degree that allows registered images is being under the situation more than the threshold value, is being compared under the big situation of the similar degree that allows registered images at the similar degree for the restriction registered images, forbid or limit the image processing of execution for input image data.
Thus, for originally not being the view data that allows registered images, but for the similar degree that allows registered images is input image data more than the threshold value, by with it as restriction registered images registered in advance, can forbid or limit the image processing of execution for this input image data.For example, although can prevent for not being the permission registered images, also be judged as the input image data that allows registered images by mistake, it is registered as the restriction registered images, thereby it is disconnected afterwards this input image data to be formed erroneous judgement, and, can prevent to allow, and the fail safe for this exception registered images is improved for the image processing of this input image data.For example, can prevent from unlawfully to carry out image processing such as duplicating, printing, transmission, conversion, editor for the exception registered images.
Other purpose of the present invention, feature and advantage can fully be understood by record shown below.And benefit of the present invention can be by understanding with reference to the following explanation of accompanying drawing.
Description of drawings
Fig. 1 is the block diagram of the schematic configuration of the document collation process unit that has in the image processing apparatus of expression an embodiment of the invention.
Fig. 2 is the block diagram of schematic configuration of the image processing apparatus of expression an embodiment of the invention.
Fig. 3 is the block diagram of the schematic configuration of the characteristic point calculation unit that has in the expression image processing apparatus shown in Figure 2.
Fig. 4 is the key diagram of the filter factor of the compound filter that has in the MTF processing unit of the image processing apparatus of expression shown in one illustration 2.
The key diagram of the center of gravity of Fig. 5 join domain that to be the characteristic point calculation unit of expression one example by image processing apparatus shown in Figure 2 extract from input image data and this join domain.
Fig. 6 is the key diagram of each center of gravity (characteristic point) of a plurality of join domains of extracting of character string that the characteristic point calculation unit of expression one example by image processing apparatus shown in Figure 2 comprises from input image data.
The concern characteristic point that Fig. 7 extracts when being the feature amount calculation unit calculated characteristics amount of expression one example by image processing apparatus shown in Figure 2 and the key diagram of peripheral characteristic point.
The key diagram of the concern characteristic point that Fig. 8 (a)～Fig. 8 (c) extracts when being the feature amount calculation unit calculated characteristics amount of expression one example by image processing apparatus shown in Figure 2 and the combination of peripheral characteristic point.
The key diagram of the concern characteristic point that Fig. 9 (a)～Fig. 9 (c) extracts when being the feature amount calculation unit calculated characteristics amount of expression one example by image processing apparatus shown in Figure 2 and the combination of peripheral characteristic point.
Figure 10 (a) and Figure 10 (b) are that expression one example is in image processing apparatus shown in Figure 2, for the hashed value of each characteristic point of registration in hash (Hash) table and the key diagram of the index of expression input image data.
Figure 11 be in the ballot processing unit of the image processing apparatus of expression shown in one illustration 2, for the chart of the votes of each registered images.
Figure 12 is the flow chart of the flow process of the processing in the expression image processing apparatus shown in Figure 2.
Figure 13 is the key diagram of the exception registered images tabulation used in the image processing apparatus of expression shown in one illustration 2.
Figure 14 (a)～Figure 14 (c) is the chart of the similar degree result of calculation in the image processing apparatus shown in expression one illustration 2.
Figure 15 (a)～Figure 15 (c) is the chart of the similar degree result of calculation in the image processing apparatus shown in expression one illustration 2.
Figure 16 is the block diagram of variation of the image processing apparatus of expression an embodiment of the invention.
Figure 17 is the key diagram of structure of the image processing system of expression an embodiment of the invention.
Figure 18 is the block diagram of a structure example of the image processing system of expression an embodiment of the invention.
Figure 19 is the block diagram of another structure example of the image processing system of expression an embodiment of the invention.
The key diagram of the concern characteristic point that Figure 20 (a)～Figure 20 (d) extracts when being the feature amount calculation unit calculated characteristics amount of expression one example by the image processing apparatus of an embodiment of the invention and the combination of peripheral characteristic point.
The key diagram of the concern characteristic point that Figure 21 (a)～Figure 21 (d) extracts when being the feature amount calculation unit calculated characteristics amount of expression one example by the image processing apparatus of an embodiment of the invention and the combination of peripheral characteristic point.
Figure 22 is the block diagram of the schematic configuration of the document collation process unit that has in the image processing apparatus of expression another embodiment of the invention.
Figure 23 (a) and Figure 23 (b) are expression one example is carried out checking object image, registered images and the voting results of collation process in document collation process unit shown in Figure 1 key diagrams.
Figure 24 is the flow chart of the flow process of the processing in the expression document collation process unit shown in Figure 22.
Embodiment
(execution mode 1)
An embodiment of the invention are described.And in the present embodiment, the situation that an example is applied the present invention to color digital compounding machine (MFP:Multi-Function Printer) describes.
(structure of 1-1. color digital compounding machine 1)
Fig. 2 is the block diagram of schematic configuration of the color digital compounding machine (image processing apparatus, image processing system, image read-out) 1 of expression present embodiment.This color digital compounding machine 1 has copy function, printing function, facsimile transmission function, scan function, is scanned up to Email (scan to e-mail) function etc.
As shown in Figure 2, color digital compounding machine 1 comprises: coloured image input unit 2, color image processing apparatus 3, coloured image output device 4, communicator 5, guidance panel 6.
For example (Charge CoupledDevice: charge coupled device) etc. optical information is transformed to scanning element (not shown) formation of the device of the signal of telecommunication to coloured image input unit (image read-out) 2 with CCD by having, will be from the reverberation picture of original copy as RGB (R: red, G: green, B: analog signal indigo plant) and output to color image processing apparatus 3.
Color image processing apparatus 3 comprises: A/D converter unit 11, shadow correction unit 12, document collation process unit 13, input gray level correcting unit 14, regional separation processing unit 15, look correcting unit 16, black undercolour removal unit 17, spatial filter processing unit 18, output gray level correcting unit 19 and the half tone reproduction processing unit 20 of generating.The analog signal that outputs to color image processing apparatus 3 from coloured image input unit 2 sends with A/D converter unit 11, shadow correction unit 12, document collation process unit 13, input gray level correcting unit 14, regional separation processing unit 15, look correcting unit 16, the black order that generates undercolour removal unit 17, spatial filter processing unit 18, output gray level correcting unit 19, half tone reproduction processing unit 20 in color image processing apparatus 3, outputs to coloured image output device 4 as the digital color-signal of CMYK.
A/D (analog/digital) converter unit 11 is transformed to digital signal with the analog signal of RGB.
The rgb signal of the numeral that 12 pairs of shadow correction unit are sent by A/D converter unit 11 implements to be used for removing the processing of the various distortions that illuminator, imaging system, the camera system of coloured image input unit 2 produce.And, the image processing system easy to handle Signal Processing that shadow correction unit 12 is implemented the adjustment of colour balance and is transformed to employings in color image processing apparatus 3 such as concentration signal.
Document collation process unit 13 carries out the homophylic judgement (similar/as similarly not judge) of input image data and registered images.And the rgb signal former state that document collation process unit 13 will be transfused to outputs to the input gray level correcting unit 14 of back level.In addition, narrate in the back for the details of document collation process unit 13.
Background color concentration) and image quality adjustment processing such as contrast input gray level correcting unit 14 has been for having removed the rgb signal of various distortions in the shadow correction unit, implements to remove background color (the concentration component of background color:.
Zone separation processing unit 15 is by rgb signal, is one of them of character zone, dot area, photo zone with each pixel separation in the input picture.Zone separation processing unit 15 is according to separating resulting, the regional identification signal that remarked pixel is belonged to which zone outputs to look correcting unit 16, black undercolour removal unit 17, spatial filter processing unit 18 and the half tone reproduction processing unit 20 of generating, and will be outputed to the look correcting unit 16 of back level by the input signal former state of input gray level correcting unit 14 outputs simultaneously.
Look correcting unit 16 is removed the turbid processing of color contamination in order to realize loyal color reproduction, and this color contamination is turbid to be based on the CMY that comprises unwanted absorption component (C: dark green, M: magenta, Y: yellow) color contamination of the dichroism of look material is turbid.
Black generate 3 chrominance signals that carry out the CMY after look is proofreaied and correct in undercolour removal unit 17 generate black (K) thus the black generation of signal, deduct the K signal that black the generation, obtains from original C MY signal and generate new CMY Signal Processing.Thus, 3 chrominance signals of CMY are transformed to 4 chrominance signals of CMYK.
Spatial filter processing unit 18 is according to regional identification signal, for being undertaken handling the correction space frequency characteristic based on the space filtering of digital filter by the black view data that generates the CMYK signal of undercolour removal unit 17 inputs.Thus, can alleviate the fuzzy and granular deterioration of output image.Half tone reproduction processing unit 20 is also the same with spatial filter processing unit 18, according to regional identification signal, the view data of CMYK signal is implemented predetermined process.
For example, be separated into the zone of character, in order to improve the reproducibility of surplus symbol or color character especially, clearly emphasize to handle the amount of emphasizing that increases high frequency by what the space filtering of spatial filter processing unit 18 was handled in regional separation processing unit 15.Simultaneously, in half tone reproduction processing unit 20, select to be suitable for binaryzation or many-valuedization processing under the high-resolution screen of reproduction of high frequency.
And, for the zone that is separated into dot area in regional separation processing unit 15, in spatial filter processing unit 18, implement to be used to remove the low-pass filtering treatment of input site component.So, after output gray level adjustment unit 19 has carried out being used for signal transformations such as concentration signals being the output gray level treatment for correcting of dot area percentage of characteristic value of coloured image output device 4, in half tone reproduction processing unit 20, implement half tone reproduction and handle (semi-tone generation), thereby this half tone reproduction is handled image finally is separated into pixel and can reproduces each gray scale.For the zone that in regional separation processing unit 15, is separated into photo, pay attention to binaryzation or many-valuedization processing under the screen of half tone reproduction.
Implemented view data that above-mentioned each handle and be temporarily stored in storage device (not shown), be read out and be input to coloured image output device 4 with predetermined timing.
Coloured image output device 4 will output to from the view data of color image processing apparatus 3 input on the recording materials (for example, paper etc.).The structure of coloured image output device 4 is not particularly limited, and for example can use the coloured image output device that has utilized electrofax mode or ink-jetting style.
Communicator 5 for example is made of modulator-demodulator or network card.Communicator 5 carries out data communication via network card, LAN cable etc. and other device that is connected to network (for example, personal computer, server unit, other digit apparatus, picture unit etc.).
And, communicator 5 is under the situation that sends view data, but thereby when carrying out guaranteeing transmit status with the process of transmitting of the other side destination, read view data with prescribed form compression (view data of reading in scanner) from memory, and implement the necessary processing such as change of compressed format, send to the other side destination successively via communication line.
And communicator 5 communicates process receiving under the situation of view data, receives the view data of sending from the other side destination simultaneously and is entered into color image processing apparatus 3.The view data that receives is implemented predetermined process such as extension process, rotation processing, conversion of resolution processing, output gray level treatment for correcting, half tone reproduction processing in color image processing apparatus 3, by 4 outputs of coloured image output device.And, also the view data that receives can be kept at storage device (not shown), and color image processing apparatus 3 is read above-mentioned data as required and is implemented the processing of afore mentioned rules.
Guidance panel 6 is for example by display unit such as LCD with set button etc. and constitute (any one is all not shown), to be presented on the above-mentioned display unit corresponding to the information of the indication of the main control unit (not shown) of color digital compounding machine 1, will be communicated to above-mentioned main control unit from the information of user's input via above-mentioned setting button simultaneously.The user can be by guidance panel 6 input for the processing requirements of input image data (for example tupe (duplicating, printing, electronics are posted a letter, the correction of facsimile transmission, filing, view data, editor etc.), handle the transmission destination of number (duplicate number, print number), input image data etc.).Above-mentioned main control unit is for example by CPU formations such as (Central Processing Unit), according to be stored in program among the not shown ROM etc. and various data, from the action of the each several part of the colored compounding machine 1 of control figures such as information of guidance panel 6 inputs.
(formation of 1-2. document collation process unit)
Then, the details to document collation process unit 13 describes.The document collation process unit 13 of present embodiment extracts a plurality of characteristic points from input image data, set to the local characteristic point of each the characteristic point decision that extracts, select the part set of characteristic point from each set of decision, with the each several part set selected amount as attendant features, according to part set in the relevant a plurality of combinations of characteristic point, ask invariant respectively for geometric conversion, make up each invariant of obtaining and calculate hashed value, by the registered images corresponding with the hashed value that calculates voted, carry out retrieval with the similar registered images of input image data, homophylic judgment processing (similar/as similarly not judge) for this registered images.
Fig. 1 is the block diagram of the schematic configuration of expression document collation process unit 13.As shown in the drawing, document collation process unit 13 comprises: characteristic point calculation unit 31, feature amount calculation unit 32, ballot processing unit 33, similar degree judgment processing unit 34, location registration process unit 35, control unit 7, memory 8.
The action of the each several part of control unit 7 control document collation process unit 13.And control unit 7 also can be included in the main control unit of the action of the each several part that is used for the colored compounding machine 1 of control figure, also can have respectively with main control unit, works in coordination with the action of controlling document collation process unit 13 with main control unit.
And, result in the similarity judgment processing of document collation process unit 13, be judged as under the situation of not similar (input image data is not the view data of registered images), control unit 7 allows to carry out image processing for input image data (processing such as duplicating, printing, electronics are posted a letter, the correction of facsimile transmission, filing, view data, editor).And under the situation that is judged as similar (input image data is the view data of registered images), control unit 7 judges whether the highest registered images of similar degree is the exception registered images.So, under the situation that is the exception registered images, allow to carry out image processing for input image data, on the other hand under the situation that is not the exception registered images, forbid or execution that limited images is handled.And, here, under the situation that is not the exception registered images, forbid or execution that limited images is handled, but be not limited thereto, for example, also can apply restriction to the additional specific pattern of image, perhaps be coated with restrictions such as bad part or all image.
Memory 8 is stored in the various data used in the processing of each several part of document collation process unit 13, result etc.And memory 8 comprises only manager (specific user of conduct; Registered in advance person with authorities such as registration, editor of threshold value that the characteristic quantity, similarity that are used for registered images and exception registered images judge etc.Safety director in enterprise, the shop etc. etc. for example) can read, edit manager's file 8A of the file (storage area) (control unit 7 can as required with reference to the information in manager's file 8A), here.
In manager's file 8A, have administrator information memory cell 101, hash table 102, exception registered images tabulation 103 etc.
101 storages of administrator information memory cell are used for determining to allow the reading of manager's file 8A, editor's gerentocratic information.Control unit 7 carries out authentication processing according to administrator information memory cell 101 under the reading of manager's file 8A, the requested situation of editor, judge whether to allow reading, editor.And the method for gerentocratic authentication processing is not particularly limited, and for example, can utilize known in the past various authentication methods.And, in the present embodiment, the manager is by the external device (ED) (computer, server unit etc.) from connecting communicatedly color digital compounding machine 1, to authenticate needed information sends to color digital compounding machine 1 and carries out authentication processing, but be not limited thereto, for example, also can carry out authentication processing via the guidance panel 6 of color digital compounding machine 1.
Hash table 102 is the tables with the characteristic quantity corresponding stored of index of registered images (ID) and registered images.Details for hash table 102 is narrated in the back.
Exception registered images tabulation 103 is tabulations of exception registered images, even this exception registered images is to judge under the situation that input image data is a registration image data, also allows the registered images of carries out image processing.Promptly, control unit 7 is being judged under the situation of view data that input image data is a registered images, though forbid or limit execution for the image processing of this input image data as principle, but at this registered images is under the situation of exception registered images, allows the execution of image processing exceptionally.
Calculate the center of gravity of coupling part the coupling part that characteristic point calculation unit 31 is extracted character string or ruling as characteristic point.Here, the structure of characteristic point calculation unit 31 is not limited thereto, and for example also can come the calculated characteristics point by known the whole bag of tricks in the past.
Fig. 3 is the block diagram of the schematic configuration of representation feature point computing unit 31.As shown in the drawing, characteristic point calculation unit 31 comprises: signal processing unit 41, resolution conversion unit 42, MTF processing unit 43, binary conversion treatment unit 44, center of gravity calculation unit 45.
Signal processing unit 41 is under the situation of coloured image in the view data (rgb signal) of 12 inputs from the shadow correction unit, with this view data netrual colourization, is transformed to lightness signal or luminance signal.
For example, signal processing unit 41 is transformed to brightness signal Y by following formula (1) with rgb signal:
Yi＝0.30Ri+0.59Gi+0.11Bi …(1)
Here, Y is the luminance signal of each pixel, and R, G, B are each color components in the rgb signal of each pixel, and adding character i is the value (i is the integer more than 1) that each pixel is given.
Perhaps, also rgb signal can be transformed to CIE1976L *a *b *Signal (CIE:CommissionInternational de l ' Eclairage; L *: lightness; a *, b *: colourity).
The 42 pairs of input image datas in resolution conversion unit become doubly to be handled.For example, becoming optically under the situation doubly by 2 pairs of input image datas of coloured image input unit, the 42 pairs of input image datas in resolution conversion unit become doubly once more, so that it becomes the resolution of regulation.And, resolution conversion unit 42 is for the treating capacity of the each processing unit of level after alleviating, also can carry out resolution conversion, the resolution of reading in when this resolution conversion makes resolution and coloured image input unit 2 grades times is compared further reduction (for example, the view data of reading in 600dppi (dot per inch) is transformed to 300dpi etc.).
(modulation transfer function: modulation transfer function) processing unit 43 is used for absorbing the spatial frequency characteristic of (adjustment) coloured image input unit 2 because of the different situation of each machine to MTF.In the picture signal of CCD output, because of the storage effect of aperture openings degree, transmission efficiency or the image retention of the sensitive surface of opticses such as lens or reflective mirror, CCD, physical scan with operate the deterioration that reason such as changeable produces MTF.Because the deterioration of this MTF, the image that reads in thickens.MTF processing unit 43 is used to repair because the processing of bluring that the deterioration of MTF produces by implementing suitable Filtering Processing (emphasizing to handle).And, unwanted high fdrequency component during the feature point extraction of the center of gravity calculation unit 45 of level is handled after also being used for being suppressed at.That is, utilize compound filter (not shown) to emphasize and smoothing processing.And Fig. 4 represents the filter factor in this compound filter of example.
Binary conversion treatment unit 44 is by the view data (brightness value (luminance signal) or brightness value (lightness signal)) of netrual colourization and pre-set threshold are compared and with the view data binaryzation.
Center of gravity calculation unit 45 is demarcated (labelling) (subsidiary mark is handled) according to the view data of having carried out binaryzation in binary conversion treatment unit 44 (for example using " 1 ", " 0 " expression) to each pixel.So, determine the join domain that the pixel of subsidiary same mark connects, with the center of gravity of the join domain determined as feature point extraction.And then, the characteristic point of extracting is outputed to feature amount calculation unit 32.The key diagram of the center of gravity of Fig. 5 join domain that to be expression one example extract from input image data and this join domain illustrates join domain corresponding with the character string of " A " and center of gravity.And Fig. 6 represents the key diagram of each center of gravity (characteristic point) of a plurality of join domains that character string that an example comprises is extracted from input image data.And above-mentioned characteristic point can be represented with the coordinate figure (x coordinate, y coordinate) in the bianry image.
Feature amount calculation unit 32 comprises feature point extraction unit 32a, invariant computing unit 32b, hashed value computing unit 32c, utilizes that the characteristic point that is calculated by characteristic point calculation unit 31 is calculated rotation with respect to original image, parallelly moves, enlarges, dwindled, the parallel characteristic quantity (hashed value and/or invariant) that geometric distortion such as moves as constant amount.
As shown in Figure 7, feature point extraction unit 32a with a characteristic point as paying close attention to characteristic point, with the characteristic point of this concerns characteristic point periphery with from the near order extraction specified quantity (being 4 points here) of the distance of paying close attention to characteristic point as peripheral characteristic point.In the example of Fig. 7, with characteristic point a when paying close attention to characteristic point, extract minutiae b, c, these 4 peripheral characteristic points of conduct of d, e, with characteristic point b when paying close attention to characteristic point, extract minutiae a, c, these 4 peripheral characteristic points of conduct of e, f.
And feature point extraction unit 32a extracts selectable combination from the peripheral characteristic point of as above extracting like that 4 at 3.For example, shown in Fig. 8 (a)～Fig. 8 (c), under with the situation of characteristic point a shown in Figure 7 as the concern characteristic point, extract 3 combination among peripheral characteristic point b, c, d, the e, be peripheral characteristic point b, c, d, peripheral characteristic point b, c, e, each combination of peripheral characteristic point b, d, e.
Then, invariant computing unit 32b is to each combination calculation of extracting invariant (one of characteristic quantity) Hij with respect to geometric distortion.Here, i is a number (i is the integer more than 1) of paying close attention to characteristic point, and j is the number (j is the integer more than 1) of the combination of 3 of the peripheral characteristic points of expression.In the present embodiment, the ratio that connects in the length of the line segment between the peripheral characteristic point two is made as invariant Hij.And the length of above-mentioned line segment also can be calculated according to each peripheral characteristic point coordinates value.For example, in the example of Fig. 8 (a), be made as A11 as the length of connection features being put the line segment of c and characteristic point d, the length of connection features being put the line segment of c and characteristic point b is made as B11, and then invariant H11 is H11=A11/B11.And, in the example of Fig. 8 (b), being made as A12 as the length of connection features being put the line segment of c and characteristic point b, the length of connection features being put the line segment of b and characteristic point e is made as B12, and then invariant H12 is H12=A12/B12.And, in the example of Fig. 8 (c), being made as A13 as the length of connection features being put the line segment of d and characteristic point b, the length of connection features being put the line segment of b and characteristic point e is made as B13, and then invariant H13 is H13=A13/B13.Like this, in the example of Fig. 8 (a)～Fig. 8 (c), calculate invariant H11, H12, H13.And, in above-mentioned example, be made as Aij though connection is positioned at the peripheral characteristic point in horizontal direction left side with the line segment that is positioned at the peripheral characteristic point of horizontal direction central authorities, the peripheral characteristic point that connection is positioned at horizontal direction central authorities is made as Bij with the line segment that is positioned at the peripheral characteristic point on horizontal direction right side, but be not limited thereto, can select the line segment that uses in the calculating of invariant Hij with arbitrary method.
Then, hashed value computing unit 32c calculates (Hi1 * 10 2+ Hi2 * 10 1+ Hi3 * 10 0The value of the remainder of)/D is stored in the memory 8 it as hashed value (one of characteristic quantity) Hi.And above-mentioned D is that what degree is the scope according to the value that can obtain remainder be set in and predefined constant.
And, the computational methods of invariant Hij are not particularly limited, for example, the value of 5 compound ratio calculating also can extracting with the configuration of the m point (m is m＜n and has removed the integer of m 〉=5) that extracts according near pay close attention to the characteristic point 5 compound ratio, extract near n point (n is the integer of n 〉=5) 5 compound ratio, near n point with from m point etc. is as the above-mentioned invariant Hij about the concern characteristic point.And so-called compound ratio is 5 values of obtaining on from the straight line 4 or the plane, is known as the invariant to a kind of projective deformation of geometric conversion.
And, the formula that is used to calculate hashed value Hi also is not limited to above-mentioned formula (2), also can use other hash function (for example, any one in the hash function of record in patent documentation 4).
And, when the each several part of feature amount calculation unit 32 finishes in the calculating of the extraction of the peripheral characteristic point of paying close attention to characteristic point for one and hashed value Hi, change to further feature point and carry out the extraction of peripheral characteristic point and the calculating of hashed value paying close attention to characteristic point, whole characteristic points are calculated hashed value.
In the example of Fig. 7, with peripheral characteristic point and the extraction of hashed value when finishing of characteristic point a when paying close attention to characteristic point, then carry out with characteristic point b when paying close attention to characteristic point peripheral characteristic point and the extraction of hashed value.In the example of Fig. 7, under with the situation of characteristic point b as the concern characteristic point, extract minutiae a, c, these 4 peripheral characteristic points of conduct of e, f.Then, shown in Fig. 9 (a)～Fig. 9 (c), extract combination (peripheral characteristic point a, e, the f that from these peripheral characteristic point a, c, e, f, select at 3, periphery characteristic point c, e, f, periphery characteristic point a, c, f), and to each combination calculation hashed value Hi, it is stored in the memory 8.Then, each characteristic point is repeated this processing, obtain respectively with each characteristic point when paying close attention to characteristic point hashed value and they are stored in the memory 8.
And, the computational methods of the invariant of characteristic point a when paying close attention to characteristic point are not limited to said method.For example, also can be shown in Figure 20 (a)～Figure 20 (d), under with the situation of characteristic point a shown in Figure 7, extract 3 combination among peripheral characteristic point b, c, d, the e, promptly as the concern characteristic point, periphery characteristic point b, c, d, periphery characteristic point b, c, e, peripheral characteristic point b, d, e, each combination of peripheral characteristic point c, d, e, to each combination of extracting, calculate invariant (one of characteristic quantity) Hij for geometric distortion.
And, with characteristic point b shown in Figure 7 when paying close attention to characteristic point, also can shown in Figure 21 (a)～Figure 21 (d), from characteristic point a, c, this peripheral characteristic point of 4 of e, f, extract certain 3 combination (peripheral characteristic point a, e, f, periphery characteristic point a, c, e, periphery characteristic point a, f, c, peripheral characteristic point e, f, c), to the invariant Hij of each combination calculation for geometric distortion.And, at this moment, calculate (Hi1 * 10 3+ Hi2 * 10 2+ Hi3 * 10 1+ Hi4 * 10 0The value of the remainder of)/D is stored in the memory 8 it and gets final product as hashed value.
And, in above-mentioned example, the line segment that connects from paying close attention to nearest peripheral characteristic point of characteristic point and the 2nd near peripheral characteristic point is made as Aij, the line segment that connects from paying close attention to nearest peripheral characteristic point of characteristic point and the 3rd near peripheral characteristic point is made as Bij, but be not limited thereto, also can be that benchmark is selected etc. with the length that connects the line segment between the peripheral characteristic point, with the selected line segment that is used to calculate invariant Hij of arbitrary method.
And feature amount calculation unit 32 will send to location registration process unit 35 for the hashed value (characteristic quantity) of each characteristic point of the input image data that as above calculates like that when carrying out the location registration process that input image data is registered as registered images.
The hashed value of each characteristic point that location registration process unit 35 will calculate for feature amount calculation unit 32, the index (original copy ID) of expression original copy (registered images) are registered to the hash table 102 (with reference to Figure 10 (a)) that is set in the memory 8 successively.Under the situation that hashed value has been registered, register original copy ID accordingly with this hashed value.Assignment numbers and original copy ID is repeated successively.And, the quantity of the original copy that is registered to hash table 102 than the many situations of setting (for example, registrable original copy quantity 80%) under, also can retrieve old original copy ID and successively with its deletion.And deleted original copy ID also can be used as the original copy ID of new registered images and reuses.And, be under the situation of identical value (H1=H5 in the example of Figure 10 (b)) in the hashed value that calculates, it is gathered be one and be registered in the hash table 102.
And, feature amount calculation unit 32 is being carried out under the situation of judgment processing (similarity judgment processing) of view data whether input image data is the registered images that has been registered, will send to ballot processing unit 33 for the hashed value of each characteristic point of the input image data of as above calculating like that.
The hashed value of each characteristic point that 33 pairs of processing units of ballot are calculated from input image data and the hashed value that is registered in the hash table 102 compare, and the registered images with same Hash value is voted.In other words, the number of times of having calculated the identical hashed value of the hashed value that has with registered images from input image data is counted, count value is stored in the memory 8 for each registered images.Figure 11 is the chart of expression one example to the votes of registered images ID1, ID2, ID3.
Ballot result (the index of each registered images and for the votes of each registered images of ballot processing unit 33 is read in similar degree judgment processing unit 34 from memory 8; Similar degree), extract maximum number of votes obtained and the index that has obtained the registered images of maximum number of votes obtained.Then, maximum number of votes obtained that relatively extracts and predetermined threshold value and judge similarity (whether input image data is the view data of registered images) send to control unit 7 with the judgement signal of expression judged result.Promptly, under maximum number of votes obtained is situation more than the predetermined threshold value, be judged as " similarity (input image data is the view data of registered images) is arranged ", and do not reaching under the situation of threshold value, be judged as " no similarity (input image data is not the view data of registered images) ".
Perhaps, similar degree judgment processing unit 34 also can be by calculating similar degree divided by ballot sum (sum of the characteristic point of extracting from input image data) back normalization for the number of votes obtained of each registered images, and by this similar degree and predetermined threshold value (for example, 80% of the ballot sum) are compared and judge similarity.
And, similar degree judgment processing unit 34 also can by will for the number of votes obtained of each registered images divided by about the registration number (registering number at most) of the hashed value of the maximum registered images of the registration number of hashed value thus similar degree is calculated in its normalization, and judge similarity by relatively this similar degree and predetermined threshold value (for example 80% of the ballot sum).That is, under the similar degree of calculating is situation more than the threshold value TH2, be judged as " similarity is arranged ", judging in the situation that does not reach threshold value TH2 gets final product into " no similarity ".And, in this case, because the sum of the hashed value of extracting from input image data is than above-mentioned maximum registration number big (particularly having the situation of hand-written part etc. at least a portion of original copy and/or registered images) sometimes, the calculated value of similar degree surpasses 100% sometimes.
And the threshold value TH2 when judging similarity both can fix each registered images, perhaps also can set each registered images according to the importance degree of each registered images etc.The importance degree of registered images also can be classified to set according to registered images, for example for bank note, marketable securities, utmost point classified papers, inner file of maintaining secrecy etc., importance degree is made as maximum, for classified document, importance degree is made as lower than bank note etc.At this moment, in memory 8, the index of the weight coefficient corresponding with the importance degree of registered images and this registered images is stored explicitly, and similar degree judgment processing unit 34 utilizes with the registered images corresponding threshold TH2 that obtains maximum number of votes obtained judges that similarity gets final product.
And, when judging similarity, also threshold value TH2 can be made as fixingly, on the other hand, the weight coefficient of each registered images be multiply by votes (number of votes obtained of each registered images) to each registered images, thereby judge similarity.At this moment, in memory 8, the index of the weight coefficient corresponding with the importance degree of each registered images and each registered images is stored explicitly, similar degree judgment processing unit 34 calculates the correction number of votes obtained that the number of votes obtained of each registered images be multiply by the weight coefficient of this registered images, proofreaies and correct number of votes obtained judgement similarity according to this and just can.For example, maximum correction number of votes obtained and threshold value TH2 can be compared, also the maximum correction number of votes obtained normalized result and threshold value TH2 comparison can be carried out with the sum of voting, also the maximum correction number of votes obtained normalized result and threshold value TH2 comparison can be carried out with maximum registration number.And at this moment, weight coefficient for example can be set at than 1 big value, and along with the raising of the importance degree of registered images and become big value.
And, in the present embodiment, a characteristic point (concern characteristic point) is calculated a hashed value, but be not limited thereto, also can calculate a plurality of hashed values to a characteristic point (concern characteristic point).For example, also can use following method,,, ask invariant to calculate the method for hashed value at 5 from 3 of 5 extractions for 6 kinds of each combinations of from these 6, extracting promptly as peripheral feature point extraction 6 points of paying close attention to characteristic point.At this moment, a characteristic point is calculated 6 hashed values.
(processing in the 1-3. color digital compounding machine 1)
Then, with reference to flow chart shown in Figure 12 processing in the color digital compounding machine 1 is described.And, in the following description, in administrator information memory cell 101, establish administrator information by the information of registered in advance.
At first, control unit 7 is obtained input image data and the user processing request (indication input) (S1, S2) via guidance panel 6 or communicator 5 inputs.And, input image data both can read original image by coloured image input unit 2 and obtain, also can obtain the input image data that sends from the device of outside, also can read and obtain input image data from various recording mediums via the card reader that has in the color digital compounding machine 1 (not shown) etc. by communicator 5.
Then, control unit 7 judges whether by above-mentioned processing request institute processing of request be location registration process (S3).Then, under the situation that is location registration process, whether the user who carries out the requirement processing is the gerentocratic authentication processing (S4) that is registered in the administrator information memory cell 101.For example, control unit 7 is obtained the user profile (user ID, password etc.) of user input via guidance panel 6 or communicator 5, by the user profile that will obtain be registered in administrator information (gerentocratic user profile) in the administrator information memory cell 101 and check and carry out authentication processing.
Then, judge authentication whether successfully (S5), judging end process under the situation that authentication failed.And, also can under the situation that authentication has been failed, allow the user to import the user profile of stipulated number again, and carry out the authentication processing of stipulated number.
On the other hand, authentication processing is under the case of successful in S4, the each several part of control unit 7 control document collation process unit 13, the input data that obtain in S1 are carried out characteristic point computing (S6) and characteristic quantity computing (S7), characteristic quantity and the original copy ID (ID of registered images) that calculates is registered to hash table 102 (S8) accordingly.
Then, control unit 7 judges by above-mentioned processing request institute processing of request it is the location registration process of registered images (forbid or limit and carry out the image of handling), the location registration process (S9) of the registered images that still makes an exception (allowing to carry out the image of processing).
Then, under the situation of location registration process that is not the exception registered images, control unit 7 end process.On the other hand, under the situation of location registration process that is the exception registered images, control unit 7 control location registration process unit 35 make the ID of this image be registered to exception registered images tabulation 103 (S10), end process.Figure 13 is the key diagram of expression one example exception registered images tabulation 103.
In S3, judge under the situation that is not location registration process, the each several part of control unit 7 control document collation process unit 13, the input data that obtain in S1 are carried out characteristic point computing (S11), and (S13), similarity judgment processing (S14) are handled in characteristic quantity computing (S12), ballot.Then, control unit 7 judges whether the result of similarity judgment processing has been judged as similar (S15).Be judged as under the no similar situation, control unit 7 allows to carry out to image processing that should input image data (duplicate, electronics is posted a letter, the correction of facsimile transmission, filing, view data, editor etc.) (S16), end process.
On the other hand, judged in S15 under the similar situation, control unit 7 judges whether the result that ballot is handled is that the highest registered images of similar degree is exception registered images (S17).That is, the result of control unit 7 by judging that ballot is handled, whether the ID of the registered images that similar degree is the highest is registered in exception registered images tabulation 103, judges whether this registered images is the exception registered images.But, be not limited thereto, for example, also can be when being registered in the characteristic quantity of registered images and original copy ID in the hash table 102, make in advance and have mark among the original copy ID (, not being that exception registered images markers is designated as " 0 " for example) for exception registered images markers is designated as " 1 ", in the processing of S17, control unit 7 is by judging whether to be the exception registered images with reference to this mark.
And control unit 7 also can be similar with being judged as in S15, is judged as the input image data (the perhaps compressing image data that this input image data is compressed) that is not the exception registered images and is stored among manager's file 8A in S17.Thus, the manager is stored in the view data among manager's file 8A by beginning from behind to read, and can confirm easily whether judged result is suitable.
On the other hand, in S17, be judged as under the situation that is not the exception registered images, forbid carrying out image processing (S18), end process input image data.On the other hand, being judged as in S15 is under the situation of exception registered images, allows to carry out the image processing (S16) to input image data, end process.
As mentioned above, in the color digital compounding machine 1 of present embodiment, the characteristic quantity of registered images (restriction registered images) of carries out image processing and the characteristic quantity of registered images (permission registered images) of permission carries out image processing are forbidden or limits in storage.And storage is used to represent that each registered images is the registered images that allows carries out image processing, still forbids or limit the information of the registered images of carries out image processing.Then, the result of the similarity judgment processing of input image data and registered images, be judged as under the similar situation, control unit 7 judges that the highest registered images of similar degree is the registered images that allows carries out image processing, still forbid or limit the registered images of carries out image processing, and carry out forbidding of image processing or limit or allow according to judged result.
Thus, can make control unit 7 registration content (whether be exception registered images) judge whether to forbid or execution that limited images is handled according to the highest registered images of similar degree.
For this point, utilize Figure 14 (a)～Figure 14 (c) and Figure 15 (a)～Figure 15 (c) to be described in detail.
Figure 14 (a)～Figure 14 (c) represents respectively as forbidding the registered images of carries out image processing and registered registered images IDa, the result's of the similar degree of input image data IDa ', IDb ' for registered images Ida during unregistered exception registered images, that calculate, IDc ' chart.And input image data IDa ' reads registered images IDa and the input image data obtained by scanner.And input image data IDb ' is to read the input image data of obtaining than higher image I Db for the registered images similar degree with scanner though be not registered images.And input image data IDc ' reads the input image data of obtaining for the lower image of the similar degree of registered images with scanner.
Shown in Figure 14 (a), input picture IDa ' is actually the view data of registered images IDa, so the similar degree height of calculating (similar degree is 100% in the example of this figure).Therefore, at this moment, input image data IDa ' is judged as the view data of registered images IDa, is under an embargo for the execution of the image processing of this input image data IDa '.
And, shown in Figure 14 (b), input image data IDb ' is the data that read than higher image from the similar degree for registered images though be not the view data of registered images in fact, so the similar degree of calculating is than higher value (being 85% in the example of this figure).Therefore, threshold value (being 80% in the example of this figure) height that the similar degree of calculating is judged than similarity although be not the view data of registered images originally, also produces the forbidden situation of execution of image processing.
And shown in Figure 14 (c), for input image data IDc ', the similar degree of calculating is lower than threshold value, and the execution of image processing is allowed to.
On the other hand, Figure 15 (a)～Figure 15 (c) is illustrated in as forbidding that the registered images of carries out image processing registered registered images IDa, and by way of exception registered images (allowing the registered images of carries out image processing) has been registered the result's of similar degree under the situation of registered images IDb, that calculate input image data IDa ', IDb ' for registered images IDa, IDc ' chart.
Shown in Figure 15 (a), input image data IDa ' is actually the view data of registered images IDa, for the similar degree also the highest (similar degree is 100% in the example of this figure) of registered images IDa.Therefore, at this moment input image data IDa ' is judged as the view data of registered images IDa, is under an embargo for the execution of the image processing of this input image data IDa '.
Shown in Figure 15 (b), input image data IDb ' is actually the view data of registered images IDb, so for the similar degree of registered images IDb the highest (being 100% in the example of this figure).And, since registered images IDb and registered images IDa originally similar degree than higher, so input image data IDb ' with respect to the similar degree of IDa also than higher (being 85% in the example of this figure).But, owing to input image data IDb ' reads the data that registered images IDb obtains, thus for the similar degree of registered images IDa unlike similar degree height for registered images IDb.Therefore, input image data IDb ' is judged as the view data of registered images IDb, allows the execution of image processing.
And shown in Figure 15 (c), for input image data IDc ', the similar degree that calculates is lower than threshold value, allows the execution of image processing.
Therefore, for example, for Figure 14 (b) although be not the view data of forbidding the registered images of carries out image processing like that originally, but be judged as the input image data (input image data IDb ') of the view data of registered images by mistake, the manager by with its by way of exception registered images register, can carry out image processing afterwards for this input image data.That is, can prevent to forbid or execution that limited images is handled owing to erroneous judgement is disconnected.
And, in the example of Figure 15 (b), surpass the threshold value that similarity is judged for the similar degree (100%) of registered images Idb with for similar degree (85%) both of registered images IDa.So, in the present embodiment, under the situation of the threshold value of judging above similarity for the similar degree of a plurality of registered images like this, the registered images the highest according to similar degree (whether the highest registered images is the exception registered images according to similar degree), whether decision needs to limit or forbid the execution of image processing.But, be not limited thereto, for example, for forbid or the similar degree of the registered images of the execution that limited images is handled under the situation more than the 2nd high threshold value of the threshold value of judging than similarity (for example 90%), perhaps for forbidding or the similar degree of the registered images of the execution that limited images is handled is under the above situation of setting (for example 90%) with respect to the peaked ratio for the similar degree of the registered images that allows carries out image processing, perhaps for the maximum of the similar degree of the registered images that allows carries out image processing with for forbidding or the difference of the similar degree of the registered images of the execution that limited images is handled does not reach under the situation such as setting (for example 5%), even for forbidding or the similar degree of the registered images of the execution that limited images is handled is lower than under the situation for the similar degree of the registered images of the execution that allows image processing, also can forbid or execution that limited images is handled.Thus, can make for forbidding or the fail safe of the registered images of the execution that limited images is handled improves.
And, in the present embodiment, even for forbidding or the similar degree of the registered images of the execution that limited images is handled is more than the threshold value judged of similarity, with respect to this similar degree, under the bigger situation of the similar degree of the exception registered images of the execution that allows image processing, all allow the execution of image processing, but be not limited thereto.For example, for forbidding or the similar degree of the registered images of the execution that limited images is handled is more than the threshold value judged of similarity, with respect to this similar degree, under the bigger situation of the similar degree of the exception registered images of the execution that allows image processing, the content that also can set the execution that allows image processing (for example, handle to allow its execution for the part among the correction of duplicating, printing, filing, view data, the editor, but handling not allow to carry out for remaining sets like that etc.).At this moment, for example, in exception registered images tabulation 103, the ID of exception registered images is stored explicitly with the information of the content of the execution of expression permission image processing, control unit 7 according to the image processing content that is required with the information of relevant exception registered images corresponding stored, decision should allow or forbid that the execution of image processing gets final product.
And, in the present embodiment,, forbid the execution handled in principle, and under the situation of this registered images for the exception registered images to having under the similar situation judging for registered images, allow the structure of the execution handled to be illustrated exceptionally.But, be not limited thereto, for example, have under the similar situation judging for registered images, the execution of processing also can be forbidden or limit to the execution of allow handling in principle (forbid under no similar situation handle execution) exceptionally under the situation of this registered images for the exception registered images.
And, in the present embodiment, the situation that applies the present invention to color digital compounding machine 1 is illustrated, but application of the present invention is not limited thereto.For example, also can be applied to monochromatic compounding machine.And, be not limited to compounding machine, for example also can be applied to facsimile communication apparatus, photocopier, image read-out of monomer etc.
Figure 16 is the block diagram of structure example that expression applies the present invention to the situation of flat bed scanner (image read-out, image processing apparatus) 1 '.
As shown in the drawing, flat bed scanner 1 ' has coloured image input unit 2 and color image processing apparatus 3 '.Color image processing apparatus 3 ' is by A/D converter unit 11, shadow correction unit 12, document collation process unit 13, control unit 7 (not shown in Figure 16), memory 8 (not shown in Figure 16) constitutes, coloured image input unit 2 is connected with it, and the composing images reading device 1 ' as a whole.And, the function of the A/D converter unit 11 in coloured image input unit (image fetching unit) 2, shadow correction unit 12, document collation process unit 13, control unit 7, memory 8 and above-mentioned color digital compounding machine 1 are roughly the same, so omit explanation here.
And, in the present embodiment, be made as manager's file 8A is included in the color digital compounding machine 1, but be not limited thereto, for example also manager's file 8A can be arranged in the external device (ED) (server unit etc.) that can be connected communicatedly with color digital compounding machine 1.
Figure 17 is the key diagram of the structure of presentation video treatment system 100, this image processing system 100 comprises: compounding machine (MFP) A, the B that has the server unit 50 of manager's file 8A, can be connected communicatedly via network and this server unit 50 ... printer A, B ... facsimile machine A, B ... computer A, B ... digital camera A, B ..., scanner A, B ... and constitute.And, the structure of image processing system 100 is not limited thereto, for example, also can be by any one the above formation in server unit 50, compounding machine, printer (image processing system), facsimile machine, computer, digital camera (image read-out), the scanner (image read-out).
Above-mentioned scanner has document board, light scanning unit, CCD (charge coupled device) etc., reads original image and generates view data by be placed on original image on the document board by light scanning unit scanning.And above-mentioned digital camera has (image-input devices) such as pick-up lens, CCD, takes original image, personage or landscape etc. and generates view data.And above-mentioned scanner and digital camera also can have the function of the image processing of implementing to stipulate for reproduced image suitably (for example various treatment for correcting etc.).Above-mentioned printer will be printed on based on the image of the view data that computer, scanner, digital camera generated on the blank (paper used for recording).And, thereby above-mentioned facsimile machine carries out processing such as binary conversion treatment, resolution conversion processing, rotation sends to the other side destination to the view data of being read in by image-input device with the view data of the form of boil down to regulation, perhaps the view data of sending from the other side destination is expanded, thereby implement according to the performance of image output device that rotation processing or resolution conversion are handled, semi-tone is handled, and as the image output of page or leaf unit.And, above-mentioned compounding machine have in scanner functions, facsimile transmission function, the printing function (copy function, printer function) more than at least two.And the aforementioned calculation machine is edited the view data of reading in by scanner or digital camera, perhaps utilizes application software to carry out making of document.
Above-mentioned each device that is connected to server unit 50 via network has control unit 7 and document collation process unit 13 respectively.So, result in the similarity judgement, judge under the situation of similar (input image data is the view data of registered images), control unit 7 is sent to server unit 50 with input image data and device ID etc. (being used for determining input image data is sent to the information of the device of server unit 50), among the control unit of server unit 50 (not shown) has the information stores that receives in server unit 50 manager's file 8A.And, judging under the homophylic situation, the user wishes to proceed to handle, and the result of user authentication process is under the situation that has allowed to proceed to handle, control unit 7 will send to server unit 50 for the sign of this input image data and device ID etc., among the control unit of server unit 50 (not shown) has the information stores that receives in server unit 50 manager's file 8A.
And, in the present embodiment, carry out the similarity judgment processing in the document collation process unit 13 that in color digital compounding machine 1, has, but be not limited thereto.For example, also can with external device (ED) that color digital compounding machine 1 can be connected communicatedly in carry out part or all function that control unit 7 and document collation process unit 13 have.
Figure 18 is the block diagram that is illustrated in the structure example under the situation that has the part of functions that control unit 7 and document collation process unit 13 had in the server unit 50.
As shown in figure 18, the color image processing apparatus 3 of color digital compounding machine 1 comprises: the memory 8a, the communicator 5 that communicates with outside device of the needed information of processing of control unit 7a, storage document collation process unit 13a of action with document collation process unit 13a, control document collation process unit 13a of characteristic point calculation unit 31 and feature amount calculation unit 32.And server unit 50 comprises: the communicator 51 that communicates with the device of outside; Document collation process unit 13b with ballot processing unit 33, similar degree judgment processing unit 34 and location registration process unit 35; The control unit 7b of control document collation process unit 13b; The memory 8b of the needed information of processing of storage document collation process unit 13b.And, when needing between each functional block that has in each functional block that in color digital compounding machine 1, has and the server unit 50 to carry out the transmission reception of data,, control unit 7a and control unit 7b control communicator 5 and 51 receive thereby suitably carrying out the transmission of data.Function for other is the same with above-mentioned structure.
And, in the example of Figure 18, whole (feature point extraction unit 32a, invariant computing unit 32b, the hashed value computing unit 32c) that in color digital compounding machine 1, have feature amount calculation unit 32, but be not limited thereto, for example also can be as shown in figure 19, in color digital compounding machine 1, have feature point extraction unit 32a and invariant computing unit 32b, on the other hand, in server unit 50, have hashed value computing unit 32c.
And, the each several part that also can in server unit 50, have feature amount calculation unit 32, the data that the characteristic point that calculates with characteristic point calculation unit 31 is relevant send to server unit 50 from color digital compounding machine 1, and the feature amount calculation unit 32 that has in the server unit 50 is calculated hashed value according to the data of hash table 102 of storing among the memory 8b and the characteristic point that receives.And, the each several part that also can in server unit 50, have characteristic point calculation unit 31 and feature amount calculation unit 32, input image data is sent to server unit 50 from color digital compounding machine 1, and the characteristic point calculation unit 31 that has in the server unit 50 and feature amount calculation unit 32 calculated hashed value according to input image data that receives from server unit 50 and the hash table 102 that is stored in the memory 8b.
And, in the above description, example to the situation of carrying out homophylic judgment processing is illustrated, but under the situation of carrying out location registration process, as long as the location registration process unit 35 that has in the server unit 50 will be registered to the hash table 102 that is arranged among the memory 8b from original copy ID and the hashed value (hashed value that the hashed value computing unit 32c that perhaps has the server unit 50 calculates) that color digital compounding machine 1 receives.And, the user that both can be color digital compounding machine 1 carries out similarity judgment processing or location registration process via guidance panel 6 appointments, and the signal that expression is carried out which kind of processing sends to server unit 50, and also can be server unit 50 be judged as no similar input picture for the result of similarity judgment processing carries out location registration process.
And, in server unit 50, have under the situation of hashed value computing unit 32c, also can be with calculating hashed value with the computational methods diverse ways that is stored in the hashed value in the hash table 102 (utilizing other hash function), and adopt the hashed value that calculates to upgrade hash table 102.Thus, for example can be according to the kind of original image etc., to hash table 102,, check precision (homophylic judgement precision) with reference to the suitable hashed value registration (renewals) of characteristic quantity (invariant) so can improve owing to its processing of voting.
And, in the present embodiment, being made as the result who judges in similarity is judging under the similar situation, input image data is stored among manager's file 8A, but, in addition also can be for example by OCR (Optical Character Recognition: optical character identification) handle the character information of storing from the input image data extraction.At this moment, for example have the OCR processing unit that extracts character information (text data) from input image data in color digital compounding machine 1, control unit 7 gets final product the text data store that this OCR extraction unit extracts in manager's file 8A.
(execution mode 2)
Another embodiment of the invention is described.And, for the convenience that illustrates, to giving identical label in execution mode, and omit its explanation with the parts that the parts that illustrate have an identical functions.
Figure 22 is the block diagram of the schematic configuration of the document collation process unit 13b that has in the color digital compounding machine 1 of expression present embodiment.And the color digital compounding machine 1 of present embodiment has the document collation process unit 13b except replacing document collation process unit 13, is the structure identical with the color digital compounding machine 1 of execution mode 1.
As shown in figure 22, document collation process unit 13b also has the virtual memory cell 9 and the log-on data delete cells 36 that are set in the memory 8 except the document collation process unit 13 of execution mode 1.
In execution mode 1, similar degree judgment processing unit 34 calculates the number of votes obtained of each registered images according to the ballot result of ballot processing unit 33, and is more than the threshold value and the highest registered images of number of votes obtained has been judged as similarity for number of votes obtained.But, only using this determination methods, can not judge rightly for the similarity of registered images sometimes.
For this point, enumerate concrete example and explain.Figure 23 (a) is checking object image α, registered images A, the B that expression comprises in the input image data, and the key diagram to the voting results of registered images A, B about checking object image α that carries out of ballot processing unit 33.And checking object image α is the image identical with registered images B.And Figure 23 (b) is checking object image α ', registered images A ', the B ' that expression comprises in the input image data, and the key diagram to the voting results of registered images A ', B ' about checking object image α ' that carries out of ballot processing unit 33.And checking object image α ' is the image identical with registered images B '.
Shown in Figure 23 (a), under the low situation of the similar degree of registered images A and B, these each registered images are relative less with respect to integral image with the same content part among the checking object image α, and different parts is more relatively.At this moment, shown in the voting results of Figure 23 (a), disturb the gained vote that is caused in the voting results for two registered images, also become greatly, therefore difficultly break to produce erroneous judgement for the difference of the number of votes obtained of different content parts even hypothesis has comprised.
On the other hand, shown in Figure 23 (b), under the high situation of the similar degree of registered images A ' and B ', these each registered images are relative more with respect to integral image with the same content part among the checking object image α ', and different parts is less relatively.At this moment, shown in the voting results of Figure 23 (b), owing to diminish for the difference of the number of votes obtained of different content parts, so comprised in the voting results for two registered images when disturbing the gained vote that is caused, it is disconnected to be easy to generate erroneous judgement.And, poor for the gained vote sum of two registered images, being difficult to discern it is to disturb produce poor, also be based on the difference of actual picture material and produce poor.
Therefore, in the present embodiment, when registration restriction registered images or permission registered images, at first carry out the virtual registration of the image of new registration, afterwards, the image of the image of the image that read in new registration with coloured image input unit 2 once more and registered registered images and above-mentioned virtual registration being carried out similar degree judges.Then, for the similar degree of registered registered images with for the difference of the similar degree of the image of virtual registration be the registration of carrying out the image of above-mentioned virtual registration under the situation more than the pre-set threshold, on the other hand, judge to discerning the registration of the image that does not carry out above-mentioned virtual registration in the situation that does not reach threshold value.
Virtual memory cell 9 is the memories with ID of the image of above-mentioned virtual registration (virtual registration ID) and characteristic quantity corresponding stored.And the characteristic quantity of the image of corresponding virtual registration both can be with itself and above-mentioned virtual registration ID corresponding stored in hash table 102, also can with above-mentioned virtual registration ID corresponding stored in the table (not shown) different with hash table 102.
Log-on data delete cells 36 is judged as under the unrecognizable situation at the image for virtual registration, the virtual registration ID and the characteristic quantity of this image of deletion from virtual memory cell 9.And, in that characteristic quantity is stored under the situation in hash table 102 or other table,, simultaneously the above-mentioned characteristic quantity of storage in hash table 102 or other tables is deleted from the virtual registration ID of virtual memory cell 9 deletions.
Figure 23 is the flow chart of the flow process of the processing among the expression document collation process unit 13b.
At first, control unit 7 and execution mode 1 the same (with reference to Figure 12), the processing requirements of obtaining input image data and importing from the user via guidance panel 6 or communicator 5 (indication input) (S1, S2) judges whether the desired processing of above-mentioned processing requirements is location registration process (S3).Then, under the situation that is not location registration process, carry out the processing of the S11～S18 among Figure 12.
On the other hand, under the situation that is location registration process, whether the user that control unit 7 carries out the requirement processing is the gerentocratic authentication processing (S21) that has been registered in the administrator information memory cell 101.For example, control unit 7 is obtained the user profile (user ID, password etc.) of user input via guidance panel 6 or communicator 5, by the user profile that will obtain be registered in administrator information (gerentocratic user profile) in the administrator information memory cell 101 and check and carry out authentication processing.
Then, judge authentication whether successfully (S22), be judged as end process under the situation that authentication failed.
On the other hand, in S21, successfully carried out under the situation of authentication processing, the each several part of control unit 7 control document collation process unit 13, the input data that obtain in S1 are carried out characteristic point computing (S23) and characteristic quantity computing (S24), will be stored in (S25) in the virtual memory cell 9 for the virtual registration ID and the characteristic quantity of this original copy.
Then, obtain the view data that has read the original copy of above-mentioned virtual registration by coloured image input unit 2 once more, the characteristic point and the characteristic quantity of this image that reads in once more carried out in calculating, carries out handling (S26) and similarity judgment processing (S27) for the registered images that is registered in the hash table 102 with the ballot that is stored in the above-mentioned virtual registered images in the virtual memory cell 9.
Then, judge the result that this similarity is judged, whether have a plurality of homophylic original copys (S28) that have been judged as.Here, do not have under a plurality of situations that are judged as homophylic original copy judging, control unit 7 carries out the later processing of S31 described later.
On the other hand, exist under a plurality of situations that are judged as homophylic original copy judging, the difference (S29) that control unit 7 calculates for the similar degree that is judged as homophylic each original copy judges whether this difference is predetermined threshold value above (S30).And, in the present embodiment, above-mentioned threshold value is made as the number of votes obtained maximum original copy similar degree 5%.
Then, in S28, judge and do not exist a plurality of being judged as to judge above-mentioned difference under the situation with homophylic original copy and in S30 under the situation more than the threshold value, control unit 7 is judged as for above-mentioned virtual registration original copy and can discerns, control location registration process unit 35 and make above-mentioned virtual registration original copy official register to hash table 102 (S31).For example, when making virtual registration ID and characteristic quantity be stored in the virtual memory cell 9, as long as the ID correspondence that this characteristic quantity and official register are used is registered in the hash table 102.And, make under virtual registration ID and the situation of characteristic quantity corresponding stored in hash table 102, as long as make the virtual registration ID of storage change to the ID that official register is used.And, also can carry out to the official register of hash table 102, control unit 7 control log-on data delete cellses 36, deletion is stored in the information in the virtual memory cell 9.
Afterwards, control unit 7 judges that by the desired processing of above-mentioned processing requirements be the registered images location registration process of (forbidding or limit the image of the execution of handling), the location registration process (S32) of the registered images that still makes an exception (allowing the image of the execution of processing).Then, under the situation of location registration process that is not the exception registered images, control unit 7 end process.On the other hand, under the situation of location registration process that is the exception registered images, control unit 7 control location registration process unit 35 make the ID of this image be stored in the tabulation 103 of exception registered images (S33) end process.
On the other hand, in S30, judge under the situation that above-mentioned difference do not reach threshold value, control unit 7 is judged as can not be discerned above-mentioned virtual registration original copy, and the information that can not register is presented on the display unit of guidance panel 6 (S34), control log-on data delete cells 36 simultaneously, deletion is stored in information in the virtual memory cell 9 and end process.
And, also can in S28, judge and have a plurality of being judged as under the situation with homophylic original copy, further judge in these a plurality of original copys, whether to comprise and allow registered images and restriction registered images, carry out the later processing of S29 under both situations comprising, on the other hand, under the situation that only comprises one of them, carry out the later processing of S31.
And, in the respective embodiments described above, constitute the file collation process unit that has in color digital compounding machine 1 and/or the server unit 50 and the each several part (each piece) of control unit, utilize processor such as CPU to realize by software.That is, color digital compounding machine 1 and/or server unit 50 have: the storage devices (recording medium) such as CPU (central processing unit), the ROM (read only memory) that has stored said procedure, the RAM (random accessmemory) that launches said procedure, storage said procedure and various memory of data etc. of carrying out the order of the control program be used to realize each function.So, purpose of the present invention reaches in the following manner, promptly, can be by computer-readable recording offer color digital compounding machine 1 and/or server unit 50 as the recording medium of the procedure code (execute form program, intermediate code program, source program) of the control program of the color digital compounding machine 1 of the software of realizing above-mentioned functions and/or server unit 50, the procedure code that is recorded on the recording medium is read and carried out to this computer (perhaps CPU or MPU).
As aforementioned recording medium, for example can use band classes such as tape or cassette tape, comprise semiconductor memory classes such as the card class such as dish class, IC-card (comprising storage card)/light-card of CDs such as disk such as floppy disk (registered trade mark)/hard disk or CD-ROM/MO/MD/DVD/CD-R or mask ROM/EPROM/EEPROM/ flash rom.
And color digital compounding machine 1 and/or server unit 50 also can constitute joinably with communication network, provide the said procedure sign indicating number via communication network.Be not particularly limited as this communication network, for example, can utilize internet, in-house network, extranets, LAN, ISDN, VAN, CATV communication network, Virtual Private Network (virtual private network), telephone wire road network, mobile radio communication, satellite communication network etc.And, be not particularly limited as the transmission medium that constitutes communication network, for example, can utilize wired modes such as IEEE1394, USB, power line transmission, wired TV circuit, telephone wire, adsl line, also can utilize wireless modes such as the such infrared ray of IrDA or remote controller, Bluetooth (registered trade mark), 802.11 wireless, HDR, portable phone net, satellite circuit, surface wave digital network.And the present invention can realize the said procedure sign indicating number in the electric transmission mode, and the form that also can pass through the computer data signal of embedding carrier wave realizes.
Each piece of color digital compounding machine 1 and/or server unit 50 is not limited to utilize software to realize, also can constitute, also can will carry out the hardware that a part handles and carry out the control that is used to carry out this hardware or the arithmetic element combination of the software of remaining processing by hardware logic.
Computer system of the present invention can be by constituting with the lower part: image-input devices such as flat bed scanner, film (film) scanner, digital camera, carry out image display device such as the computer of various processing such as above-mentioned similar degree computing or similarity judgment processing, the CRT monitor that shows the result of computer, LCD by the program of lading clearance and the image processing systems such as printer that the result of computer outputs to paper etc. are constituted.And, can also have as the network card of the means of communication that is used for being connected or modulator-demodulator etc. via network and server etc.
As mentioned above, in order to solve above-mentioned problem, image processing apparatus of the present invention comprises: the input data obtain the unit, obtain input image data; The Characteristic Extraction unit extracts the characteristic quantity of above-mentioned input image data; Stored registered images characteristic quantity memory cell with obtain from the registered images of being obtained the characteristic quantity of registered images by the external device (ED) that can be connected communicatedly the unit at least one; The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of registered images, thereby calculates the similar degree of two images; And control unit, whether according to above-mentioned similar degree is more than the predetermined threshold value, whether decision will forbid or limit the image processing of carrying out for above-mentioned input image data, it is characterized in that, in above-mentioned registered images, permission registered images as the registered images that allows carries out image processing is arranged, with as forbidding or limit the restriction registered images of the registered images of carries out image processing, above-mentioned image processing apparatus also has identifying information and obtains the unit, this identifying information is obtained the unit and is obtained the identifying information that is used to discern above-mentioned permission registered images and above-mentioned restriction registered images from said memory cells or said external device, above-mentioned control unit is under the maximum of the similar degree that above-mentioned similar degree computing unit is calculated is situation more than the above-mentioned threshold value, at similar degree is that peaked registered images is to allow under the situation of registered images, allow to carry out image processing for above-mentioned input image data, at similar degree is that peaked registered images is under the situation of restriction registered images, forbids or limit execution for the image processing of above-mentioned input image data.
And, above-mentioned image processing is not limited to the image processing of narrow senses such as the distortion, conversion, coding of image, for example comprises: duplicating, printing, electronics are posted a letter, correction of facsimile transmission, filing, view data, editor, conversion, coding etc. be for all processing of view data.And, above-mentioned input data obtain the unit for example both can obtain input image data by reading original copy by scanner, also can obtain input image data by utilizing software that the form of necessary item input electronic data is made, also can obtain the input image data that directly makes as electronic data, also can obtain the input image data that sends from the device that can connect communicatedly image processing apparatus.
According to above-mentioned structure, the Characteristic Extraction unit extracts the characteristic quantity that the input data obtain the input image data of obtaining the unit.Then, thus the characteristic quantity of the input image data that the similar degree computing unit extracts the Characteristic Extraction unit compares the similar degree that calculates two images with the characteristic quantity of registered images or the registered images characteristic quantity of obtaining the registered images of obtaining from the external device (ED) that can be connected communicatedly the unit in being stored in memory cell.And, in above-mentioned registered images, have as the permission registered images of the registered images that allows carries out image processing with as the restriction registered images of forbidding or limit the registered images of carries out image processing.Then, identifying information is obtained the unit and is obtained the identifying information that is used to discern permission registered images and restriction registered images from memory cell or external device (ED), control unit is under the maximum of the similar degree that the similar degree computing unit is calculated is situation more than the above-mentioned threshold value, at similar degree is that peaked registered images is to allow under the situation of registered images, allow to carry out image processing for input image data, and at similar degree is that peaked registered images is under the situation of restriction registered images, forbids or limit execution for the image processing of input image data.
Thus, because registration has the registered images of permission and restriction registered images, so, under the maximum of the similar degree that the similar degree computing unit calculates is situation more than the above-mentioned threshold value, can be to allow registered images or restriction registered images according to the registered images of similar degree maximum, carries out image processing will be forbidden or limit to decision whether.Therefore, although for example can prevent that also being broken by erroneous judgement is the input image data of restriction registered images for not being the restriction registered images, it as allowing registered images to register, is broken thereby afterwards this input image data is formed to judge by accident.And, in contrast,, it is registered as the restriction registered images, thereby it is disconnected afterwards this input image data to be formed erroneous judgement although can prevent from also to be judged as the input image data that allows registered images by mistake for not being to allow registered images.
Image processing apparatus of the present invention can be following structure also, promptly comprises: the input data obtain the unit, obtain input image data; The Characteristic Extraction unit extracts the characteristic quantity of above-mentioned input image data; Stored forbid or limit carries out image processing the restriction registered images characteristic quantity memory cell with obtain from the registered images of the characteristic quantity of being obtained the restriction registered images by the external device (ED) that can be connected communicatedly the unit at least one; The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of restriction registered images, thereby calculates the similar degree of two images; And control unit, under above-mentioned similar degree is situation more than the predetermined threshold value, forbid or limit the image processing of carrying out for this input image data, it is characterized in that, above-mentioned image processing apparatus also has the location registration process unit, this location registration process unit makes the characteristic quantity of the permission registered images that allows carries out image processing, be stored in the 2nd memory cell that said memory cells or said external device have with the identifying information that is used for discerning above-mentioned restriction registered images and above-mentioned permission registered images, above-mentioned similar degree computing unit calculates the similar degree of above-mentioned input image data for above-mentioned restriction registered images, with the similar degree of above-mentioned input image data for above-mentioned permission registered images, above-mentioned control unit is more than the above-mentioned threshold value at the similar degree for above-mentioned restriction registered images, and be compared under the big situation of the similar degree of above-mentioned restriction registered images for the similar degree of above-mentioned permission registered images, allow to carry out image processing for above-mentioned input image data.
According to said structure, the Characteristic Extraction unit extracts the characteristic quantity that the input data obtain the obtained input image data in unit.Then, the characteristic quantity of the input image data that similar degree computing unit comparative feature amount extraction unit extracts and the characteristic quantity of the restriction registered images that is stored in memory cell or registered images is obtained the characteristic quantity of the restriction registered images of obtaining from the external device (ED) that can be connected communicatedly the unit, thereby calculate the similar degree of two images, control unit is forbidden or is limited the image processing of carrying out for this input image data under above-mentioned similar degree is situation more than the predetermined threshold value.And, image processing apparatus also has the location registration process unit, this location registration process unit makes the characteristic quantity of the permission registered images that allows carries out image processing and is used for discerning the restriction registered images and allows the identifying information of registered images to be stored in the 2nd memory cell that memory cell or external device (ED) have, similar degree computing unit calculating input image data for the similar degree of restriction registered images and input image data for the similar degree that allows registered images.Then, control unit is more than the threshold value and is compared to for the similar degree that allows registered images under the big situation of the similar degree of restriction registered images at the similar degree for the restriction registered images, allows to carry out the image processing for input image data.
Thus, for the view data that originally is not the restriction registered images, but the similar degree for the restriction registered images is a input image data more than the threshold value, by with it as allowing the registered images registered in advance, can carry out image processing for this input image data.For example, although can prevent for not being the restriction registered images, also by the disconnected input image data of erroneous judgement for the restriction registered images, it is registered as the permission registered images, thereby it is disconnected afterwards this input image data to be formed erroneous judgement, and, can prevent to be forbidden irrelevantly or limit for the image processing of this input image data.
And image processing apparatus of the present invention also can be used as following structure, promptly comprises: the input data obtain the unit, obtain input image data; The Characteristic Extraction unit extracts the characteristic quantity of above-mentioned input image data; Memory cell and the registered images of obtaining the characteristic quantity that allows registered images from the external device (ED) that can be connected communicatedly of having stored the characteristic quantity of the permission registered images that allows carries out image processing obtained at least one the unit; The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of permission registered images, thereby calculates the similar degree of two images; And control unit, under above-mentioned similar degree is situation more than the predetermined threshold value, allow to carry out image processing for this input image data, it is characterized in that, above-mentioned image processing apparatus also has the location registration process unit, this location registration process unit makes the characteristic quantity of forbidding or limiting the restriction registered images of carries out image processing, be stored in the 2nd memory cell that said memory cells or said external device have with the identifying information that is used for discerning above-mentioned permission registered images and above-mentioned restriction registered images, above-mentioned similar degree computing unit calculates the similar degree of above-mentioned input image data for above-mentioned permission registered images, with the similar degree of above-mentioned input image data for above-mentioned restriction registered images, above-mentioned control unit is more than the above-mentioned threshold value at the similar degree for above-mentioned permission registered images, and be compared under the big situation of the similar degree of above-mentioned permission registered images for the similar degree of above-mentioned restriction registered images, forbid or limit the image processing of carrying out for above-mentioned input image data.
According to said structure, the Characteristic Extraction unit extracts the characteristic quantity that the input data obtain the input image data of obtaining the unit.Then, the characteristic quantity of the input image data that similar degree computing unit comparative feature amount extraction unit extracts be stored in memory cell in the characteristic quantity of permission registered images or the registered images characteristic quantity of obtaining the permission registered images of obtaining from the external device (ED) that can be connected communicatedly the unit, thereby calculate the similar degree of two images, control unit allows to carry out the image processing for this input image data under similar degree is situation more than the predetermined threshold value.And, image processing apparatus also has the location registration process unit, this location registration process unit makes the characteristic quantity and being used for of forbidding or limiting the restriction registered images of carries out image processing limit registered images and allows the identifying information of registered images to be stored in the 2nd memory cell that memory cell or external device (ED) have, and similar degree computing unit calculating input image data are for the similar degree that allows registered images and the input image data similar degree for the restriction registered images.Then, control unit is being to be compared to more than the threshold value and for the similar degree of restriction registered images under the big situation of the similar degree that allows registered images for the similar degree that allows registered images, forbids or limit the image processing of execution for input image data.
Thus, for originally not being the view data that allows registered images, but for the similar degree that allows registered images is input image data more than the threshold value, by with it as restriction registered images registered in advance, can forbid or limit the image processing of execution for this input image data.For example, although can prevent for not being the permission registered images, also be judged as the input image data that allows registered images by mistake, it is registered as the restriction registered images, thereby it is disconnected afterwards this input image data to be formed erroneous judgement, and, can prevent to allow, and the fail safe for this exception registered images is improved for the image processing of this input image data.For example, can prevent from unlawfully to carry out image processing such as duplicating, printing, transmission, conversion, editor for the exception registered images.
And, image processing apparatus also can be used as following structure, promptly, the similar degree that above-mentioned similar degree computing unit calculates be peaked registered images be allow registered images, be under the situation more than the 2nd threshold value bigger for the similar degree of restriction registered images than above-mentioned threshold value, be more than above-mentioned the 2nd threshold value for the similar degree of restriction registered images perhaps and above-mentioned maximum and be under the situation below the setting for the difference of the similar degree of above-mentioned restriction registered images, above-mentioned control unit is forbidden or limit the image processing of execution for above-mentioned input image data.
According to said structure, even the registered images in the similar degree maximum is to allow under the situation of registered images, under the similar degree for the restriction registered images is situation more than the 2nd threshold value bigger than above-mentioned threshold value, be more than above-mentioned the 2nd threshold value and for the similar degree that allows registered images with for the difference of the similar degree of above-mentioned restriction registered images to be under the situation below the setting perhaps, forbid or limit the image processing of execution for above-mentioned input image data for the similar degree of restriction registered images.Thus, can improve for the fail safe that limits registered images.
And, image processing apparatus also can be used as following structure, promptly, above-mentioned image processing apparatus has virtual log-on message memory cell, this virtual log-on message cell stores is to the information as the registry object image that is required the image registered, above-mentioned control unit is when registering above-mentioned registry object image as above-mentioned registered images, carry out following processing: give virtual registration ID to above-mentioned registry object image, make above-mentioned virtual registration ID simultaneously and be stored in the processing of above-mentioned virtual log-on message memory cell from the characteristic quantity that this registry object image extracts; Obtain with image fetching unit read in the original copy corresponding once more with above-mentioned registry object image and obtain read in image again, according to reading in the characteristic quantity that image extracts again from this, calculating is for the similar degree of above-mentioned registry object image of storing in the above-mentioned virtual log-on message memory cell and registered registered images in said memory cells or above-mentioned the 2nd memory cell, and to judge whether to exist a plurality of similar degrees be the above treatment of picture of predetermined threshold value; Be judged as under the situation that does not have a plurality of above-mentioned images, making above-mentioned registry object image formally be registered in processing in said memory cells or above-mentioned the 2nd memory cell as registered images by above-mentioned location registration process unit; And be judged as under the situation that has a plurality of above-mentioned images, calculating is poor for the above-mentioned similar degree of these a plurality of images, in this difference is under the situation more than the setting, by above-mentioned location registration process unit above-mentioned registry object image formally is registered in said memory cells or above-mentioned the 2nd memory cell as registered images, on the other hand, do not reaching under the situation of setting, forbidding of the processing of above-mentioned registry object image official register in said memory cells or above-mentioned the 2nd memory cell.
According to said structure, under the situation of the registration that has required the registry object image, judge whether to discern this registry object image and registered intact registered images, and under discernible situation, register, on the other hand, under situation about can not discern, can forbid the registration of this registry object image.Therefore, can prevent and to be registered again with the image of registered intact registered images identification, so can judge input picture and restriction registered images accurately and allow which of registered images similar.
Image processing apparatus also can be used as following structure, promptly, above-mentioned image processing apparatus has the log-on data delete cells, this log-on data delete cells does not reach in above-mentioned difference under the situation of setting, and deletion is about being stored in the information of the above-mentioned registry object image in the above-mentioned virtual log-on message memory cell.
According to above-mentioned structure, therefore the information owing to deleting from virtual log-on message memory cell about the registry object image that can not discern, can prevent the leakage about the information of this registry object image.
Image processing system of the present invention comprises: above-mentioned any one image processing apparatus; And will be formed on image output unit on the recording materials corresponding to the image of input image data.
According to above-mentioned image processing system, because registration restriction registered images and permission registered images, so under the maximum of the similar degree that the similar degree computing unit calculates is situation more than the above-mentioned threshold value, can be to allow registered images or the restriction registered images determines whether forbidding or to limit carries out image processing according to the registered images of similar degree maximum.
In order to solve above-mentioned problem, image processing method of the present invention comprises: the input data obtain operation, obtain input image data; The Characteristic Extraction operation is extracted the characteristic quantity of above-mentioned input image data; The log-on data that obtains the characteristic quantity of registered images obtains operation; The similar degree calculation process compares the characteristic quantity of input image data and the characteristic quantity of registered images, thereby calculates the similar degree of two images; And treatment limits operation, whether according to above-mentioned similar degree is more than the predetermined threshold value, whether decision will forbid or limit the image processing of carrying out for above-mentioned input image data, it is characterized in that, in above-mentioned registered images, permission registered images as the registered images that allows carries out image processing is arranged, with as forbidding or limit the restriction registered images of the registered images of carries out image processing, above-mentioned image processing method also has identifying information and obtains operation, the operation that obtains this identifying information obtains and is used to discern the identifying information that allows registered images and restriction registered images, in above-mentioned treatment limits operation, under the maximum of the similar degree that above-mentioned similar degree calculation process is calculated is situation more than the above-mentioned threshold value, at similar degree is that peaked registered images is to allow under the situation of registered images, allow to carry out image processing for above-mentioned input image data, at similar degree is that peaked registered images is under the situation of restriction registered images, forbids or limit execution for the image processing of above-mentioned input image data.
According to above-mentioned method, be extracted in the characteristic quantity of importing the input image data of obtaining in the data extract operation in the Characteristic Extraction operation.Then, in the similar degree calculation process, thereby compare the similar degree that calculates two images to the characteristic quantity of the input image data that in the Characteristic Extraction operation, extracts with at the characteristic quantity that log-on data obtains the registered images of obtaining in the operation.And, in above-mentioned registered images, comprise as the permission registered images of the registered images that allows carries out image processing with as the restriction registered images of forbidding or limit the registered images of carries out image processing.Then, obtain in the operation at identifying information, obtain the identifying information that is used to discern permission registered images and restriction registered images from memory cell or external device (ED), in the treatment limits operation, the maximum of the similar degree of calculating in the similar degree calculation process is under the situation more than the above-mentioned threshold value, at similar degree is that peaked registered images is to allow under the situation of registered images, allow to carry out image processing for input image data, and at similar degree is that peaked registered images is under the situation of restriction registered images, forbids or limit execution for the image processing of input image data.
Thus, under the maximum of the similar degree that the similar degree calculation process calculates is situation more than the above-mentioned threshold value, can determine whether according to the registered images of similar degree maximum forbidding or to limit carries out image processing.Therefore, although for example can prevent that also being broken by erroneous judgement is the input image data of restriction registered images for not being the restriction registered images, it as allowing registered images to register, is broken thereby afterwards this input image data is formed to judge by accident.And, in contrast,, it is registered as the restriction registered images, thereby it is disconnected afterwards this input image data to be formed erroneous judgement although can prevent from also to be judged as the input image data that allows registered images by mistake for not being to allow registered images.
In order to solve above-mentioned problem, image processing system of the present invention comprises: image processing apparatus; And for the server unit that this image processing apparatus can connect communicatedly, disperse to have in above-mentioned image processing apparatus and above-mentioned server unit: the input data obtain the unit, obtain input image data; Memory cell has been stored the characteristic quantity of registered images; The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of registered images, thereby calculates the similar degree of two images; And control unit, whether according to above-mentioned similar degree is more than the predetermined threshold value, whether decision will forbid or limit the image processing of carrying out for above-mentioned input image data, it is characterized in that, in above-mentioned registered images, permission registered images as the registered images that allows carries out image processing is arranged, with as forbidding or limit the restriction registered images of the registered images of carries out image processing, above-mentioned image processing system also has identifying information and obtains the unit, this identifying information is obtained the unit and is obtained the identifying information that is used to discern above-mentioned permission registered images and above-mentioned restriction registered images from said memory cells, above-mentioned control unit is under the maximum of the similar degree that above-mentioned similar degree computing unit is calculated is situation more than the above-mentioned threshold value, at similar degree is that peaked registered images is to allow under the situation of registered images, allow to carry out image processing for above-mentioned input image data, at similar degree is that peaked registered images is under the situation of restriction registered images, forbids or limit execution for the image processing of above-mentioned input image data.
According to above-mentioned structure, the Characteristic Extraction unit extracts the characteristic quantity that the input data obtain the input image data of obtaining the unit.Then, thus the characteristic quantity of the input image data that the similar degree computing unit extracts the Characteristic Extraction unit and the characteristic quantity that is stored in the registered images in the memory cell compare the similar degree that calculates two images.And, in above-mentioned registered images, have as the permission registered images of the registered images that allows carries out image processing with as the restriction registered images of forbidding or limit the registered images of carries out image processing.Then, identifying information is obtained the unit and is obtained the identifying information that is used to discern permission registered images and restriction registered images from memory cell or external device (ED), control unit is under the maximum of the similar degree that the similar degree computing unit is calculated is situation more than the above-mentioned threshold value, at similar degree is that peaked registered images is to allow under the situation of registered images, allow to carry out image processing for input image data, and at similar degree is that peaked registered images is under the situation of restriction registered images, forbids or limit execution for the image processing of input image data.
Thus, because registration has the registered images of permission and restriction registered images as registered images, so, under the maximum of the similar degree that the similar degree computing unit calculates is situation more than the above-mentioned threshold value, can be to allow registered images or restriction registered images according to the registered images of similar degree maximum, carries out image processing will be forbidden or limit to decision whether.Therefore, although for example can prevent that also being broken by erroneous judgement is the input image data of restriction registered images for not being the restriction registered images, it as allowing registered images to register, is broken thereby afterwards this input image data is formed to judge by accident.And, in contrast,, it is registered as the restriction registered images, thereby it is disconnected afterwards this input image data to be formed erroneous judgement although can prevent from also to be judged as the input image data that allows registered images by mistake for not being to allow registered images.
And, above-mentioned image processing apparatus also can pass through computer realization, at this moment, by making computer carry out work, thereby be also contained in the category of the present invention by the image processing program of the above-mentioned image processing apparatus of computer realization and the recording medium that writes down the embodied on computer readable of this program as each part mentioned above.
Concrete or the embodiment that finishes in the detailed explanation item of invention understands technology contents of the present invention, should only not be defined in such object lesson and carry out the explanation of narrow sense, in the scope of the claim item of spirit of the present invention and following record, can carry out various changes and implement.

Claims (9)

1. image processing apparatus comprises:
The input data obtain the unit, obtain input image data;
The Characteristic Extraction unit extracts the characteristic quantity of described input image data;
Stored registered images characteristic quantity memory cell with obtain from the registered images of being obtained the characteristic quantity of registered images by the external device (ED) that can be connected communicatedly the unit at least one;
The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of registered images, thereby calculates the similar degree of two images; And
Whether control unit is more than the predetermined threshold value according to described similar degree, and whether decision will forbid or limit the image processing of carrying out for described input image data, it is characterized in that,
In described registered images, have as the permission registered images of the registered images that allows carries out image processing with as the restriction registered images of forbidding or limit the registered images of carries out image processing,
Described image processing apparatus also has identifying information and obtains the unit, and this identifying information is obtained the unit and obtained the identifying information that is used to discern described permission registered images and described restriction registered images from described memory cell or described external device (ED),
Described control unit
Under the maximum of the similar degree that described similar degree computing unit is calculated is situation more than the described predetermined threshold value,
At similar degree is that peaked registered images is under the situation of restriction registered images, forbid or limit execution for the image processing of described input image data,
Similar degree be peaked registered images be allow registered images, for the similar degree of restriction registered images for the situation that does not reach 2nd threshold value bigger than described threshold value under, allow to carry out image processing for described input image data,
At similar degree is that peaked registered images is to allow registered images, (i) be under the situation more than described the 2nd threshold value at similar degree for the restriction registered images, perhaps, be more than described the 2nd threshold value and described maximum and be under the situation below the setting for the difference between the similar degree of described restriction registered images (ii), forbid or limit the image processing of execution for described input image data at similar degree for the restriction registered images.
2. image processing apparatus as claimed in claim 1 is characterized in that,
Described image processing apparatus has virtual log-on message memory cell, and this virtual log-on message cell stores is to the information as the registry object image that is required the image registered,
Described control unit carries out following processing when registering described registry object image as described registered images:
Give virtual registration ID to described registry object image, make described virtual registration ID simultaneously and be stored in the processing of described virtual log-on message memory cell from the characteristic quantity that this registry object image extracts;
Obtain with image fetching unit read in the original copy corresponding once more with described registry object image and obtain read in image again, according to reading in the characteristic quantity that image extracts again from this, calculating is for the described registry object image of storing in the described virtual log-on message memory cell and the similar degree of registered registered images in described memory cell, and to judge whether to exist a plurality of similar degrees be the above treatment of picture of predetermined threshold value;
Be judged as under the situation that does not have a plurality of described images, making described registry object image formally be registered in processing in the described memory cell as registered images by described location registration process unit; And
Be judged as under the situation that has a plurality of described images, calculating is poor for the described similar degree of these a plurality of images, in this difference is under the situation more than the setting, by described location registration process unit described registry object image formally is registered in the described memory cell as registered images, on the other hand, do not reaching under the situation of setting, forbidding of the processing of described registry object image official register to described memory cell.
3. image processing apparatus comprises:
The input data obtain the unit, obtain input image data;
The Characteristic Extraction unit extracts the characteristic quantity of described input image data;
Stored forbid or limit carries out image processing the restriction registered images characteristic quantity memory cell with obtain from the registered images of the characteristic quantity of being obtained the restriction registered images by the external device (ED) that can be connected communicatedly the unit at least one;
The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of restriction registered images, thereby calculates the similar degree of two images; And
Whether control unit is more than the predetermined threshold value according to described similar degree, and whether decision is forbidden or limited the image processing of carrying out for described input image data, it is characterized in that,
Described image processing apparatus also has the location registration process unit, the identifying information that this location registration process unit makes the characteristic quantity of the permission registered images that allows carries out image processing and is used for discerning described restriction registered images and described permission registered images is stored in the 2nd memory cell that described memory cell or described external device (ED) have
Described similar degree computing unit calculates described input image data for the similar degree of described restriction registered images and the described input image data similar degree for described permission registered images,
Described control unit
Under the maximum of the similar degree that described similar degree computing unit is calculated is situation more than the described predetermined threshold value,
At similar degree is that peaked registered images is under the situation of restriction registered images, forbid or limit execution for the image processing of described input image data,
Similar degree be peaked registered images be allow registered images, for the similar degree of restriction registered images for the situation that does not reach 2nd threshold value bigger than described threshold value under, allow to carry out image processing for described input image data,
At similar degree is that peaked registered images is to allow registered images, (i) be under the situation more than described the 2nd threshold value at similar degree for the restriction registered images, perhaps, be more than described the 2nd threshold value and described maximum and be under the situation below the setting for the difference between the similar degree of described restriction registered images (ii), forbid or limit the image processing of execution for described input image data at similar degree for the restriction registered images.
4. image processing apparatus comprises:
The input data obtain the unit, obtain input image data;
The Characteristic Extraction unit extracts the characteristic quantity of described input image data;
Memory cell and the registered images of obtaining the characteristic quantity that allows registered images from the external device (ED) that can be connected communicatedly of having stored the characteristic quantity of the permission registered images that allows carries out image processing obtained at least one the unit;
The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of permission registered images, thereby calculates the similar degree of two images; And
Whether control unit is more than the predetermined threshold value according to described similar degree, and whether decision allows to carry out the image processing for described input image data, it is characterized in that,
Described image processing apparatus also has the location registration process unit, the identifying information that this location registration process unit makes the characteristic quantity and being used for of forbidding or limiting the restriction registered images of carries out image processing discern described permission registered images and described restriction registered images is stored in the 2nd memory cell that described memory cell or described external device (ED) have
Described similar degree computing unit calculates described input image data for the similar degree of described permission registered images and the described input image data similar degree for described restriction registered images,
Described control unit
Under the maximum of the similar degree that described similar degree computing unit is calculated is situation more than the described predetermined threshold value,
At similar degree is that peaked registered images is under the situation of restriction registered images, forbid or limit execution for the image processing of described input image data,
Similar degree be peaked registered images be allow registered images, for the similar degree of restriction registered images for the situation that does not reach 2nd threshold value bigger than described threshold value under, allow to carry out image processing for described input image data,
At similar degree is that peaked registered images is to allow registered images, (i) be under the situation more than described the 2nd threshold value at similar degree for the restriction registered images, perhaps, be more than described the 2nd threshold value and described maximum and be under the situation below the setting for the difference between the similar degree of described restriction registered images (ii), forbid or limit the image processing of execution for described input image data at similar degree for the restriction registered images.
5. as claim 3 or 4 described image processing apparatus, it is characterized in that,
Described image processing apparatus has virtual log-on message memory cell, and this virtual log-on message cell stores is to the information as the registry object image that is required the image registered,
Described control unit carries out following processing when registering described registry object image as described registered images:
Give virtual registration ID to described registry object image, make described virtual registration ID simultaneously and be stored in the processing of described virtual log-on message memory cell from the characteristic quantity that this registry object image extracts;
Obtain with image fetching unit read in the original copy corresponding once more with described registry object image and obtain read in image again, according to reading in the characteristic quantity that image extracts again from this, calculating is for the similar degree of described registry object image of storing in the described virtual log-on message memory cell and registered registered images in described memory cell or described the 2nd memory cell, and to judge whether to exist a plurality of similar degrees be the above treatment of picture of predetermined threshold value;
Be judged as under the situation that does not have a plurality of described images, making described registry object image formally be registered in processing in described memory cell or described the 2nd memory cell as registered images by described location registration process unit; And
Be judged as under the situation that has a plurality of described images, calculating is poor for the described similar degree of these a plurality of images, in this difference is under the situation more than the setting, by described location registration process unit described registry object image formally is registered in described memory cell or described the 2nd memory cell as registered images, on the other hand, do not reaching under the situation of setting, forbidding of the processing of described registry object image official register to described memory cell or described the 2nd memory cell.
6. image processing apparatus as claimed in claim 5 is characterized in that,
Described image processing apparatus has the log-on data delete cells, and this log-on data delete cells does not reach in described difference under the situation of setting, and deletion is about being stored in the information of the described registry object image in the described virtual log-on message memory cell.
7. an image processing system is characterized in that, comprising:
The described image processing apparatus of claim 1; And
To be formed on the image output unit on the recording materials corresponding to the image of input image data.
8. image processing method comprises:
The input data obtain operation, obtain input image data;
The Characteristic Extraction operation, the characteristic quantity of extraction input image data;
The log-on data that obtains the characteristic quantity of registered images obtains operation;
The similar degree calculation process compares the characteristic quantity of input image data and the characteristic quantity of registered images, thereby calculates the similar degree of two images; And
Whether the treatment limits operation is more than the predetermined threshold value according to described similar degree, and whether decision will forbid or limit the image processing of carrying out for described input image data, it is characterized in that,
In described registered images, have as the permission registered images of the registered images that allows carries out image processing with as the restriction registered images of forbidding or limit the registered images of carries out image processing,
Described image processing method also has identifying information and obtains operation, and this identifying information is obtained operation and obtained the identifying information that is used to discern permission registered images and restriction registered images,
In described treatment limits operation,
Under the maximum of the similar degree that described similar degree calculation process is calculated is situation more than the described predetermined threshold value,
At similar degree is that peaked registered images is to allow under the situation of registered images, forbid or limit the image processing of execution for described input image data,
Similar degree be peaked registered images be allow registered images, for the similar degree of restriction registered images for the situation that does not reach 2nd threshold value bigger than described threshold value under, allow to carry out image processing for described input image data,
At similar degree is that peaked registered images is to allow registered images, (i) be under the situation more than described the 2nd threshold value at similar degree for the restriction registered images, perhaps, be more than described the 2nd threshold value and described maximum and be under the situation below the setting for the difference between the similar degree of described restriction registered images (ii), forbid or limit the image processing of execution for described input image data at similar degree for the restriction registered images.
9. an image processing system comprises: image processing apparatus; And the server unit that can connect communicatedly for this image processing apparatus,
In described image processing apparatus and described server unit, disperse to have:
The input data obtain the unit, obtain input image data;
Memory cell has been stored the characteristic quantity of registered images;
The similar degree computing unit compares the characteristic quantity of input image data and the characteristic quantity of registered images, thereby calculates the similar degree of two images; And
Whether control unit is more than the predetermined threshold value according to described similar degree, and whether decision will forbid or limit the image processing of carrying out for described input image data, it is characterized in that,
In described registered images, have as the permission registered images of the registered images that allows carries out image processing with as the restriction registered images of forbidding or limit the registered images of carries out image processing,
Described image processing system also has identifying information and obtains the unit, and this identifying information is obtained the unit must be used to discern the identifying information of described permission registered images and described restriction registered images from described cell array area,
Described control unit
Under the maximum of the similar degree that described similar degree computing unit is calculated is situation more than the described predetermined threshold value,
At similar degree is that peaked registered images is under the situation of restriction registered images, forbid or limit execution for the image processing of described input image data,
Similar degree be peaked registered images be allow registered images, for the similar degree of restriction registered images for the situation that does not reach 2nd threshold value bigger than described threshold value under, allow to carry out image processing for described input image data,
At similar degree is that peaked registered images is to allow registered images, (i) be under the situation more than described the 2nd threshold value at similar degree for the restriction registered images, perhaps, be more than described the 2nd threshold value and described maximum and be under the situation below the setting for the difference between the similar degree of described restriction registered images (ii), forbid or limit the image processing of execution for described input image data at similar degree for the restriction registered images.
CN2008100900071A 2007-03-28 2008-03-27 Image processing apparatus, image forming apparatus, image processing system, and image processing method Expired - Fee Related CN101277368B (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
JP2007085926 2007-03-28
JP85926/07 2007-03-28
JP58611/08 2008-03-07
JP2008058611A JP4336729B2 (en) 2007-03-28 2008-03-07 Image processing apparatus, image forming apparatus, image transmitting apparatus, image reading apparatus, image processing system, image processing program, and recording medium thereof

Publications (2)

Publication Number Publication Date
CN101277368A CN101277368A (en) 2008-10-01
CN101277368B true CN101277368B (en) 2011-04-06

Family

ID=39996349

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2008100900071A Expired - Fee Related CN101277368B (en) 2007-03-28 2008-03-27 Image processing apparatus, image forming apparatus, image processing system, and image processing method

Country Status (2)

Country Link
JP (1) JP4336729B2 (en)
CN (1) CN101277368B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012514814A (en) * 2009-01-06 2012-06-28 シーメンス・ヘルスケア・ダイアグノスティックス・インコーポレーテッド Method and apparatus for automatic detection of the presence and type of caps on vials and other containers
WO2010092952A1 (en) * 2009-02-10 2010-08-19 公立大学法人大阪府立大学 Pattern recognition device
WO2010106644A1 (en) * 2009-03-17 2010-09-23 富士通株式会社 Data collating device and program
JP2012191368A (en) * 2011-03-10 2012-10-04 Sharp Corp Image processing apparatus and method of operating the same
KR101180471B1 (en) 2011-09-27 2012-09-07 (주)올라웍스 Method, apparatus and computer-readable recording medium for managing reference face database to improve efficiency of face recognition using small-capacity memory
JP6011185B2 (en) * 2012-09-14 2016-10-19 株式会社バッファロー Image information processing system, image information processing apparatus, and program
JP6151141B2 (en) * 2013-09-18 2017-06-21 株式会社東芝 Sorting device and sorting method
US10574848B1 (en) * 2018-10-26 2020-02-25 Toshiba Tec Kabushiki Kaisha Information processing apparatus with registered image detection

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1494037A (en) * 2002-11-01 2004-05-05 ��ʽ���綫֥ Apparatus and method for identifying personnel and passage controller
JP2004334337A (en) * 2003-04-30 2004-11-25 Canon Inc Image processor
CN1752870A (en) * 2004-09-22 2006-03-29 夏普株式会社 Image processing method and apparatus

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1494037A (en) * 2002-11-01 2004-05-05 ��ʽ���綫֥ Apparatus and method for identifying personnel and passage controller
JP2004334337A (en) * 2003-04-30 2004-11-25 Canon Inc Image processor
CN1752870A (en) * 2004-09-22 2006-03-29 夏普株式会社 Image processing method and apparatus

Also Published As

Publication number Publication date
JP2008269582A (en) 2008-11-06
CN101277368A (en) 2008-10-01
JP4336729B2 (en) 2009-09-30

Similar Documents

Publication Publication Date Title
CN101388073B (en) Image checking device, image checking method and image data input processing device
CN101431582B (en) Image processing apparatus, image forming apparatus, image processing system, and image processing method
CN101277368B (en) Image processing apparatus, image forming apparatus, image processing system, and image processing method
CN101382770B (en) Image matching apparatus, image matching method, and image data output processing apparatus
CN101398649B (en) Image data output processing apparatus and image data output processing method
CN101571698B (en) Method for matching images, image matching device, image data output apparatus, and recording medium
CN101184137B (en) Image processing method and device, image reading and forming device
CN101286200B (en) Image processing apparatus and methhod, image forming apparatus, and image processing system
CN101277367B (en) Image processing apparatus, image forming apparatus, image processing system, and image processing method
CN101360160B (en) Image processing apparatus, image processing system, image processing method, and image forming apparatus
CN101320426B (en) Image processing device and method, image forming device and image processing system
CN101505349B (en) Image processing method, image processing apparatus and system, image reading apparatus and image forming apparatus
JP4378413B2 (en) Image processing apparatus, image forming apparatus, image reading apparatus, image processing system, image processing method, image processing program, and recording medium thereof
CN101404020B (en) Image processing method, image processing apparatus, image forming apparatus, image reading apparatus
CN101320425B (en) Image processing apparatus, image forming apparatus, and image processing method
CN101369314B (en) Image processing apparatus, image forming apparatus, image processing system, and image processing method
CN101364268B (en) Image processing apparatus and image processing method
CN101520846B (en) Image processing method, image processing apparatus and image forming apparatus
CN101393414B (en) Image data output processing apparatus and image data output processing method
CN101150647A (en) Image processing device, image forming device and image processing system
CN101354717B (en) Document extracting method and document extracting apparatus

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20110406

Termination date: 20210327