CN101261682A - Image processing apparatus, image processing method, and computer program product - Google Patents
Image processing apparatus, image processing method, and computer program product Download PDFInfo
- Publication number
- CN101261682A CN101261682A CNA2008100825564A CN200810082556A CN101261682A CN 101261682 A CN101261682 A CN 101261682A CN A2008100825564 A CNA2008100825564 A CN A2008100825564A CN 200810082556 A CN200810082556 A CN 200810082556A CN 101261682 A CN101261682 A CN 101261682A
- Authority
- CN
- China
- Prior art keywords
- view data
- image
- image processing
- processing apparatus
- characteristic feature
- 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.)
- Granted
Links
Images
Landscapes
- Facsimile Image Signal Circuits (AREA)
- Image Analysis (AREA)
- Editing Of Facsimile Originals (AREA)
- Character Input (AREA)
Abstract
The invention provides an image processing apparatus, an image processing method and a computer program product thereof. The image processing apparatus includes an image acquiring unit that acquires image data, a characteristic-feature acquiring unit that acquires a characteristic feature of the image data based on pixel value distribution in the image data, a determining unit that determines whether the image data corresponds to captured image data based on the characteristic feature, and an image processing unit that performs image processing on the image data depending on the result of determination obtained by the determining unit.
Description
Relevant cross reference
The application requires at the Japanese priority file 2007-053978 of application on March 5th, 2007 with in the right of priority of the Japanese priority file 2007-325145 of application on Dec 17th, 2007, and full content is included in here as a reference.
Technical field
The present invention relates to a kind of technology according to the view data carries out image processing.
Background technology
Image processing apparatus is carried out the Flame Image Process to various images.An example of image to be processed be by the image read-out of for example scanner or digital camera etc. by from manual documentation etc. optically reads image data obtain image and view data (being called " catching image " hereinafter) analoged to numeral (A/D) conversion.The example of Flame Image Process comprises that scale-of-two is handled and GTG is handled.
Another example of image to be processed be by will be for example the text that produces of the display screen of the main frame of personal computer or main frame be converted to the image that view data (being called " real electronic image " hereinafter) obtains.Real electronic image need not to carry out the A/D conversion and the electronization generation.
Image processing apparatus is a Flame Image Process and to operation computer program operation parameter.Usually come setup parameter corresponding to the resolution (counting on the per inch) of view data to be processed.Therefore, comprise information about resolution if catch the view data of image and real electronic image, then image processing apparatus can accurately be carried out the Flame Image Process that character for example extracts based on resolution.
Comprise the information about resolution in some view data, and do not comprise the information about resolution in other view data, this is to depend on the view data how to obtain.
When obtaining by scanner when catching image, catch image and obtain information about resolution by reading by scanner.Yet,, only obtain about the information of pixel and can not obtain information about resolution when obtaining by digital camera when catching image.
About real electronic image, the information that need not to set about resolution just can obtain real electronic image in image processing apparatus.Therefore, as the image of catching that is obtained by digital camera, real electronic image does not comprise the information about resolution sometimes yet.
If view data does not comprise the information about resolution, image processing apparatus carries out image processing correctly then.Under these circumstances, image processing apparatus uses the default resolution of setting in its operating system to come carries out image processing.
Yet, may stand image degradation by the view data of using default resolution processes.For example, font size (view data size) meeting changes along with the default resolution of the pixel of foundation view data, and therefore view data can not be processed as the user is desired.
Further, to catching image and to real electronic image carries out image processing in a different manner.Especially,, noise whether visible by the image on the back side of considering when the scanning to look from the surface that will scan at paper or gamma characteristic come to carry out scale-of-two and handle catching image.On the other hand, above-mentioned consideration is not necessarily for real electronic image.Therefore, if by using computer program identical and identical parameter to come, handle then that degree of accuracy can reduce and processing speed can reduce to real electronic image carries out image processing with use in catching image.
Japanese Laid-Open Patent Application No.2003-271897 discloses a kind of character recognition device.This character recognition device is from by the pixel of obtaining character in the character string that reads the view data, and the font size of the character that obtains by hypothesis be standard (for example, 10.5 pound) come calculating resolution, thereby character recognition device can come setup parameter based on the resolution of calculating.As a result, can come execution character to extract by stationary mode.
Above-mentioned character recognition device can be to not comprising the suitable resolution of view data setting about the information of resolution; Yet, for by using different computer programs and different parameter to carry out different Flame Image Process and be still difficulty to catching image and real electronic image.
The operator can catch image or real electronic image according to the view data of input and come image processing apparatus is set computer program and the parameter that is fit to.Yet, be very loaded down with trivial details and reluctant for all carrying out above-mentioned setting when the each carries out image processing, and each all to carry out above-mentioned setting in suitable mode also be difficult according to the type of image.
Summary of the invention
The objective of the invention is to the problem of small part solution in conventional art.
According to an aspect of the present invention, provide a kind of image processing apparatus.This image processing apparatus comprises: image acquisition unit is used to obtain view data; The feature acquiring unit is used to obtain the characteristic feature of view data; Determining unit is used for determining based on characteristic feature the type of view data; And graphics processing unit, be used for type according to view data to the view data carries out image processing.
According to a further aspect in the invention, provide a kind of image processing method.This image processing method comprises the following steps: to receive view data; Obtain the characteristic feature of view data; Determine the type of view data based on characteristic feature; And according to the type of view data to the view data carries out image processing.
In accordance with a further aspect of the present invention, provide a kind of computer program that is used for carrying out on computers said method.
When considering with reference to the accompanying drawings, followingly can be better understood for above and other objects of the present invention, feature, advantage and technology and industrial significance for detailed description of the currently preferred embodiment of the present invention by reading.
Description of drawings
Fig. 1 is the block scheme according to the image processing apparatus of first embodiment of the invention;
Fig. 2 is the functional block diagram of image processing apparatus as shown in Figure 1;
The serve as reasons process flow diagram of the Flame Image Process that as shown in Figure 1 image processing apparatus carries out of Fig. 3;
Fig. 4 is the histogram that is used to illustrate according to the view data of the background area of first embodiment;
Fig. 5 determines the detail flowchart handled based on the view data of as shown in Figure 3 background area;
Fig. 6 is the functional-block diagram according to the image processing apparatus of second embodiment of the invention;
Fig. 7 is the histogram that is used to illustrate the view data of the target area of handling according to second embodiment;
Fig. 8 is a process flow diagram of determining processing based on the view data of the color cluster of being carried out by image processing apparatus as shown in Figure 6;
Fig. 9 is the process flow diagram that image processing apparatus is as shown in Figure 6 handled the view data execution gray scale of input;
Figure 10 carries out the process flow diagram that scale-of-two is handled for image processing apparatus as shown in Figure 6 based on the character color that obtains from document files;
Figure 11 carries out the process flow diagram that scale-of-two is handled for image processing apparatus as shown in Figure 6 based on the character color that uses color cluster to obtain;
Figure 12 carries out the process flow diagram that crooked correction is handled for image processing apparatus as shown in Figure 6 to the view data of input;
Figure 13 is the functional-block diagram according to the image processing apparatus of third embodiment of the invention;
The view data that Figure 14 carries out based on distributing in the regular line zone of view data interior pixel value for image processing apparatus is as shown in figure 13 determined the process flow diagram of processing;
Figure 15 is the functional-block diagram according to the image processing apparatus of fourth embodiment of the invention;
The example of the binary picture data that Figure 16 obtains for as shown in figure 15 image processing apparatus;
Figure 17 is the enlarged diagram in part rule line zone as shown in figure 16;
The view data that Figure 18 carries out based on the modified example of the width in the regular line zone of view data for image processing apparatus is as shown in figure 15 determined the process flow diagram of processing; And
Figure 19 is the functional-block diagram according to the image processing apparatus of fifth embodiment of the invention.
Embodiment
Below by describing exemplary embodiment of the present invention with reference to the accompanying drawings in detail.
Fig. 1 is the block scheme according to the image processing apparatus 1 of first embodiment of the invention.Image processing apparatus 1 comprises scanner 10, CPU (central processing unit) (CPU) 11, random-access memory (ram) 12, ROM (read-only memory) (ROM) 13, storer 14, Zip disk (CD)-ROM/FD (floppy disk) driver 15, printing equipment 16, display device 17 and facsimile recorder 18.
In following embodiment, by carrying out Flame Image Process is described to catching image and real electronic image.Optical imagery reading device by for example scanner or digital camera comes the optically read image of catching.The optical imagery reading device does not read real electronic image after image generates.The example of real electronic image comprises text image data and the view data that is generated by predetermined application, and certain the format-pattern data after text image data or the conversion of above-mentioned image data format.
Scanner 10 optically read text or images about source document, and be view data with text or the image transitions that reads.The view data that reads is converted into monochromatic image data (black white image) or color graphics data (multi-level images) again, and it is stored in the middle of the storer 14.Optically read in the above described manner and image that be stored in the storer is considered to catch image.
Storer 14 is stored the view data that the view data that is read by scanner 10 and the processing of being obtained by carries out image processing obtain temporarily therein.
CD-ROM/FD driver 15 and printing equipment 16 read the computer program of being carried out by CPU 11 from CD-ROM or FD.
In case receive print command, printing equipment 16 is print image data on paper and analog just.
The setting or the state of display device 17 display image treating apparatus 1.Display device 17 is the view data that obtains of display process suitably also.
Facsimile recorder 18 will be stored in view data in the storer 14 etc. by telephone wire 19 and be sent to the external image treating apparatus.
Scanner 10, printing equipment 16 and facsimile recorder 18 are all integrated to be installed in the middle of the image processing apparatus 1; Yet they also can link to each other with image processing apparatus 1 individually by network.
If view data with GIF (EXIF) storage, then by in advance the label increase expression of EXIF being caught the information of image or real electronic image, is determined the type of view data.
Suppose the text image data execution character is extracted processing, if view data for catching image, is then carried out by optical character identification (OCR) and handled to obtain text message.On the other hand, if view data is real electronic image, then by from original electron information (for example, at the text data that is converted between the view data) rather than utilize and obtain text message among the OCR and carry out processing more accurately.
Fig. 2 is the functional-block diagram of image processing apparatus 1.Image processing apparatus 1 comprises image acquisition unit 201, regional determining unit 202, characteristic feature (characteristic-feature) acquiring unit 203, determining unit 204 and graphics processing unit 205.
The input of image acquisition unit 201 receiving target view data.The example of destination image data is included in the view data of storage in the storer 14 and the view data that receives by the network (not shown).
This be since when reading images optically and with the data that read when simulated data is converted to numerical data, this A/D conversion can cause picture noise.For example, when background color is white in original image, if owing to the characteristic feature of pixel or the optically read picture noise that occurs of image, then the RGB of background color (red, green, blueness) value can become non-unified value (i.e. " FFF4EF " or " FAD9FA ") from unified value (i.e. " FFFFFF ").Further, when optically read paper, image processing apparatus 1 is understood the image on the back side of paper that scanning is not target image.At large, catch image and have aforesaid characteristic feature.Therefore, image processing apparatus 1 is defined as target area to be processed with the background area, and determines that by obtaining characteristic feature whether original image is for catching image.
Characteristic feature acquiring unit 203 obtains the characteristic feature of remarked pixel value distribution (variation) from view data.Especially, characteristic feature acquiring unit 203 distribution of in the zone of determining by regional determining unit 202, obtaining pixel value.
Determining unit 204 determines that based on the characteristic feature of obtaining (distribution value) image that reads catches image or real electronic image.Especially, when the distribution value of obtaining during greater than preset threshold in ROM 13, determining unit 204 determines that original images are for catching image.On the other hand, when the distribution value of obtaining was equal to or less than threshold value, determining unit 204 determined that original image is real electronic image.
As mentioned above, image processing apparatus 1 is that view data or the real electronic image of importing selected appropriate image processing based on view data, and carries out the Flame Image Process of choosing.
Fig. 3 is the process flow diagram by the Flame Image Process of image processing apparatus 1 execution.
As mentioned above, (for example carrying out the predetermined picture processing, gray scale is handled or character extract handle) time, image processing apparatus 1 is determined the type of the view data of input, thereby and carries out appropriate image processing with pinpoint accuracy or carries out image processing at a high speed based on definite result.
Will describe a kind of image processing apparatus 1 below and how determine that the view data of importing catches the image or the method for real electronic image.
Real electronic image has the characteristic feature of pixel value unified in white portion.That is to say that if the zone of view data is white, then the whole pixel values in this zone are (R, G, B)=(255,255,255).
On the other hand, when scanner 10 reading images are when white (catching image), because the characteristic feature or the picture noise of the element in scanner 10, (R, G, B)=(255,255,255) that may not be that whole pixel values of image are.Therefore, for example the pixel value in such big zone, background area is not unified usually in catching image.
Determine the background area by utilization disclosed prior art in disclosed Japanese patent application No.2001-297303, and the threshold value that storage in advance is used to determine in ROM 13.
Can also determine the background area by the regional as a setting mode in the zone in the preset range of the peak value of supposition as shown in the histogram of view data.Fig. 4 is the histogram that is used for determining from peak value the view data of background area.In histogram shown in Figure 4, the zone with peak brightness is confirmed as the background area.Especially, the zone in the preset range of peak value is confirmed as the background area.Subsequently, the pixel value based on the background area distributes to determine that whether image is for catching image.Also can determine the background area by other method.
Fig. 5 is the definite detail flowchart of handling of the view data in step S302 execution as shown in Figure 3.
In case determine the background area, characteristic feature acquiring unit 203 is scanning color pixel (step S503) in the background area just, and calculates the pixel value distribution (step S504) of background area.
Determining unit 204 compared pixels values distribute and pre-set threshold (step S505) in image processing apparatus 1.When pixel value distributes greater than threshold value (step S505 is), because the pixel value disunity in the background area, so determining unit 204 just determines that view data is for catching image (step S506).On the other hand, when the pixel value distribution is equal to or less than threshold value (step S505 denys), because the pixel value in the background area is unified, so determining unit 204 just determines that view data are real electronic image (step S507).
The definite in the above described manner image of image processing apparatus 1 is caught image or real electronic image.Correspondingly, catch image or real electronic image according to image, image processing apparatus 1 makes graphics processing unit 205 carry out suitably Flame Image Process.As a result, carries out image processing more accurately.
Other factors outside also can distributing based on the pixel value in the background area of view data determines that view data catches image or real electronic image.Next will be described in the second embodiment of the invention that color cluster (clustering) wherein is used to determine the type of view data.
Fig. 6 is the functional-block diagram according to the image processing apparatus 600 of second embodiment.Image processing apparatus 600 comprises image acquisition unit 201, color cluster unit 601, characteristic feature acquiring unit 602, determining unit 603 and graphics processing unit 205.Different with image processing apparatus 1, image processing apparatus 600 does not comprise regional determining unit 202.The operation of characteristic feature acquiring unit 602 and determining unit 603 is also different with determining unit 204 with characteristic feature acquiring unit 203.Same parts is endowed and reference number identical in first embodiment, and omits the explanation for them.
601 pairs of color cluster unit have been carried out the view data of Flame Image Process and have been carried out color cluster, and form color cluster.The example of color cluster comprises the color cluster of background area and text filed color cluster.Color cluster is the method that a kind of each scope about pixel value is come grouping (grouping) color in view data.
Characteristic feature acquiring unit 602 obtains the characteristic feature of the pixel value distribution of presentation video data.Especially, characteristic feature acquiring unit 602 obtains the pixel value distribution about each color cluster that is formed by color cluster unit 601.
Based on the characteristic feature (distribution value) of each color cluster, determining unit 603 definite view data are caught image or real electronic image.
When optically read, catch image and can cause picture noise.In order to detect this picture noise, must determine in the target area in the image of catching that detects picture noise.The background area is used as the target area in first embodiment; Yet picture noise also may appear at other zone (content area) of catching image.Therefore, according to second embodiment, not only background area but also content area all are assumed that the target area.That is to say that the pixel value that is based on each color cluster (for example, the color of the presumptive area of the color of background area, text filed color, photo) distributes to determine that view data catches image or real electronic image.Therefore, even when the color of background area in gradient the time, or even after from the background area, removing picture noise, also can carry out and above-mentionedly determine.
When removing fict electronic image and comprising the gradient image of photo for example or color grade, real electronic image forms with a limited number of color usually.The example of real electronic image comprises the text image that obtains from text data.
On the other hand, catch image and form with many colors usually, and comprise the image of high gradient.For example, when catching image when being text image, even the same character on line also is a gradient color.In other words, some part is darker than the other parts color of character, and forms the border between character and the background area in the blend color of the color in the color and background zone of character.
As mentioned above, color cluster is carried out in color cluster unit 601, characteristic feature acquiring unit 602 calculates the number of the color cluster in view data and the pixel value of each color cluster distributes, and determining unit 603 determines that the distribution value that calculates is greater than threshold value or is equal to or less than threshold value.As a result, can determine that image catches image or real electronic image.Especially, when distribution value during greater than threshold value, because therefore big distribution value representation, determines that image is to catch image as the existence of the number of color of the characteristic feature of catching image.On the other hand, when the distribution value is equal to or less than threshold value,, determine that therefore image is real electronic image in image because little distribution value representation is limited as the number of color that uses of the characteristic feature of real electronic image.
Determining unit 603 is carried out deterministic process by use the threshold value of storing in advance in ROM 13.Alternatively, can be worth according to the distribution of the number of cluster and each cluster and set suitable threshold.
Color cluster is carried out by using prior art in color cluster unit 601.Fig. 7 is the histogram that is used to illustrate the view data in the zone of being handled by image processing apparatus 600.As can be seen from Figure 7, when forming a plurality of color cluster by color cluster, for color cluster a plurality of peak values have appearred respectively.Therefore, a plurality of zones in the brightness preset range of peak value are considered to zone to be processed.Each regional calculating pixel value to be processed is distributed, thereby can distribute to determine that image catches image or real electronic image based on pixel value.
Fig. 8 is a process flow diagram of determining processing based on the view data of the color cluster of being carried out by image processing apparatus 600.
When the distribution value of all clusters during all greater than threshold value (step S804 is), determining unit 603 determines that images are for catching image (step S805).When the distribution value one of them is equal to or less than threshold value at least the time (step S804 not), determining unit 603 determines that images are real electronic image (step S806).
Therefore, image processing apparatus 600 by processing as shown in Figure 8 determine the input view data catch image or real electronic image.As a result, need not for the user to consider that the type (view data is caught image or real electronic image) of view data just can be to the image processing apparatus input image data.
The Flame Image Process of being carried out by image processing apparatus 600 according to the type of view data is described below.Image processing apparatus 1 also can be carried out identical processing.
The example of Flame Image Process comprises that gray scale is handled, scale-of-two is handled, character extracts and handles and edge treated.Flame Image Process can be carried out in many ways.Especially, in the middle of gray scale was handled, the brightness of the pixel value by calculating each pixel in multi-level images (redness within one of them in 256 tones of each color, green, blue pixels value) was converted to multi-level images the gray level image of 256 tones (brightness) from white to black.
A method calculating brightness is to calculate brightness aforesaid redness from each pixel, green, the blue pixels value.Alternatively, can use the pixel value of green of each pixel as brightness.Preceding a kind of method needs more time; Yet, can obtain the image of pinpoint accuracy.A kind of method in back has realized the high speed image processing; Yet, when the raw image data handled in limited color, be difficult to obtain the image of pinpoint accuracy.
As mentioned above, carries out image processing in several ways.Real electronic image comprises pure color, for example the pure green of being represented by (R, G, B)=(0,255,0).Be used as the brightness of entire image as the pixel value of fruit green, the part in the then pure green can be become white by Flame Image Process.Therefore preferably, from red, green, blue pixels value, calculate brightness for each pixel in the real electronic image.
On the other hand, although green pixel value is used as the brightness of catching image, but since catch image comprise a plurality of gradient color also the degree of accuracy of the image of speed up processing and processing can not degenerate a lot.Therefore preferably, use green pixel value as the brightness of catching image.
Therefore according to the type of view data to be processed, appropriate image processing is selected and carried out to image processing apparatus 600.
Fig. 9 carries out the process flow diagram that gray scale is handled by the view data of 600 pairs of inputs of image processing apparatus.
When view data is (step S902 is) when catching image, graphics processing unit 205 is converted to gray level image by high speed processing with view data, that is to say that the pixel value of the green by using each pixel is converted to gray level image (step S903) as brightness with view data.
On the other hand, when view data is real electronic image (step S902 denys), graphics processing unit 205 is converted to gray level image (step S904) by calculating brightness for each pixel of real electronic image from each red, green, blue pixel value with view data.
Equally, can carry out scale-of-two to the view data of text image in several ways handles.
The example that scale-of-two is handled comprises wherein each pixel value of view data and the processing of threshold.The pixel that is equal to or less than threshold value is converted into white and is converted into black to form the black and white monochrome image greater than the pixel of threshold value.
In the text image of real electronic image, same color has identical pixel value.For example, the character of black is represented uniformly by (R, G, B)=(0,0,0).Therefore, by obtaining in advance about the information of character color to come effectively real electronic image to be converted to gray level image, and be black and other color pixel is converted to white with the pixel transitions of character color.
Can obtain information by using several different methods about character color.For example, by utilizing prior art to obtain information from the position that text (text data before being converted into view data) obtains character in advance about character color.Carry out subsequently color cluster with same color classification for about the group of each pixel value, and the group that determines to have the small pixel value with respect to text.
Figure 10 and Figure 11 are the process flow diagram of being handled by the scale-of-two that image processing apparatus 600 is carried out.In the middle of the threshold value that each processing is used all is stored in ROM 13.
Handle based on the scale-of-two that the information about character color that obtains from document files is carried out as shown in figure 10.The process of carrying out at step S1001 and S1002 is identical with the process of step S901 as shown in Figure 9 and S902 execution; Therefore, omitted explanation for them.
When the view data of determining input (step S1002 is) when catching image, thereby relatively the pixel value and the threshold value of each pixel in view data are converted to binary picture (step S1003) with view data to graphics processing unit 205.
When the view data of determining input is real electronic image (step S1002 denys), graphics processing unit 205 never is converted to the position (step S1004) that obtains character in the text of view data, and the pixel value of the position that acquires, i.e. character color (step S1005).Based on the character color by character color being converted to black and other color conversion being obtained for white, graphics processing unit 205 is converted to view data binary picture (step S1006) subsequently.
The information about character color that obtains based on utilizing the color cluster of being carried out by graphics processing unit 205, the scale-of-two of carrying out is as shown in figure 11 handled.
The process of carrying out at step S1101 and S1102 is identical with the process of carrying out at step S901 and S902, so has omitted the explanation to them.
When the view data of determining input (step S1102 is) when catching image, thereby relatively the pixel value and the threshold value of each pixel in view data are converted to binary picture (step S1103) with view data to graphics processing unit 205.
When the view data of determining input is real electronic image (step S1102 denys), color clusters (step S1104) are carried out in color cluster unit 601, and determine character color (step S1105) from the result and the color distribution of color cluster.Based on the character color by character color being converted to black and other color conversion being obtained for white, graphics processing unit 205 is converted to view data binary picture (step S1106) subsequently.
In the processing of describing about Figure 10 and Figure 11, when the view data of input is real electronic image, by character color being converted to black and being that white is converted to binary picture with view data with other color conversion.Yet, text (character) part can be converted to first color and other parts is converted to second color.In other words, can view data be converted to binary picture based on textual portions and other parts.
Crooked in the corrected image data must be according to the value of the type change parameter of view data.
Catch image owing to reading by scanner 10 and may comprising crooked.For example, when the document with image that is read by scanner 10 is set at the read direction of scanner 10 with being tilted, can cause catching in the image occur crooked.If appearance is crooked in catching image, then need to be repaired.
On the other hand, electronization forms real electronic image, and therefore, can not occur by document set and cause crooked.Therefore, when the view data that is read by image processing apparatus 600 was real electronic image, the skew angle that automatic setting need be corrected was zero.That is to say, do not carry out crooked correction.
The process flow diagram that Figure 12 handles for the crooked correction of carrying out by the view data of 205 pairs of inputs of graphics processing unit.Can carry out crooked correction in first and second embodiment handles.
The process of carrying out in step S1201 and S1202 is identical with process performed in step S901 shown in Figure 9 and S902; Therefore omitted explanation for them.
When the view data of determining input is (step S1202 is) when catching image, graphics processing unit 205 is crooked obtaining skew angle θ (step S1203) by using prior art to detect, and corrects crooked θ (step S1205).
On the other hand, when the view data of determining input is real electronic image (step S1202 denys), graphics processing unit 205 automatic setting skew angle θ are zero (step S1204), promptly do not carry out crooked correction.
As mentioned above, image processing apparatus 600 is determined the type of the view data of input, and according to Flame Image Process such as the type of the view data of input for example gray scale is handled to carry out to mode and pinpoint accuracy efficiently, scale-of-two processing and crooked correction.
Carry out as the process described in Fig. 3, Fig. 5 and Fig. 8 to Figure 12 by image processing program.Image processing program for example is stored among the ROM 13, and is carried out on image processing apparatus 1 by CPU 11.
According to first and second embodiment, distribute to determine that based on the pixel value in definite zone of the background area of for example view data view data catches image or real electronic image.Alternatively, also can based on by the optically read of view data characteristic feature that cause carry out and above-mentionedly determine.The third embodiment of the invention that the type of view data is determined in wherein rule-based line (ruled-line) zone is below described.
Figure 13 is the functional-block diagram according to the image processing apparatus 1300 of the 3rd embodiment.Image processing apparatus 1300 comprises image acquisition unit 201, binary unit 1301, regular line zone determining unit 1302, characteristic feature acquiring unit 1303, determining unit 1304 and graphics processing unit 205.The operation of characteristic feature acquiring unit 1303 and determining unit 1304 is different with the operation of characteristic feature acquiring unit 203 and determining unit 204.Same parts is endowed the reference number identical with the parts of image processing apparatus 1, and has omitted the explanation for them.
Thereby the view data of 1301 pairs of inputs of binary unit is carried out binarization and is handled the generation binary picture data.
Rule line zone determining unit 1302 is determined regular line zone from binary picture data.Can determine regular line zone by using several different methods.For example, from view data, extract and equal threshold value or, and the group that extracts and coupled another group black picture element combine than a group black picture element on the long spanning length of threshold value.The group that extracts combination subsequently is as solid regular line.Equal predetermined threshold or the working length longer if the solid regular line that extracts has, then discern it as regular line than predetermined threshold.
Characteristic feature acquiring unit 1303 obtains pixel value and distributes from binary picture data.Especially, characteristic feature acquiring unit 1303 calculates and is distributing corresponding to the pixel value in the zone (view data before being converted into binary picture) in the regular line zone of being determined by regular line zone determining unit 1302.
Usually be formed on the regular line that form in the general file and square use with monochrome.In other words, seldom come the formation rule line with polychrome or gradient color.For real electronic image, because regular line is same color, therefore the pixel value of above-mentioned regular line distributes and is considered to zero.On the other hand, for catching image, because optically read, therefore the color change with regular line is gradient color or similar color.For example, when wanting optically read paper media, be expressed in color on the paper media by point usually, and the zone of roughly the same color comprises multiple color for the object that will print.When by the optically read above-mentioned paper media of scanner, because high-resolution ground optics reading images and degree of accuracy ground explanation point more cause the big distribution of the color in the target area.Owing to also above-mentioned color change can occur by the optically read picture noise that causes.
As mentioned above, can think that occurring pixel value hardly in the regular line zone in the binary picture of real electronic image distributes.On the other hand, can think that occurring pixel value in catching image usually distributes.According to the 3rd embodiment, characteristic feature acquiring unit 1303 calculates and is distributing corresponding to the pixel value in the input picture zone in the regular line zone of binary picture.Therefore, catching image in the time of can determining image still is real electronic image.
Based on the distribution value in the regular line zone of being calculated by characteristic feature acquiring unit 1303, determining unit 1304 definite images are caught image or real electronic image.Especially, equal or during greater than predetermined threshold, determining unit 1304 determines that images are to catch image when the distribution value.
If obtained many regular lines, then determining unit 1304 is carried out definite by using several different methods.An example is that determining unit 1304 is carried out said process in proper order to the regular line that each bar obtains, and determines that immediately image is to catch image after having greater than the regular line of the color distribution value of predetermined threshold detecting.Subsequently, process is not handled the regular line of remainder detecting end constantly.Another example is that 1304 pairs of whole regular lines of determining unit are carried out said process, and when the number that has greater than the regular line that outnumbers distribution value of the regular line of the distribution value of threshold value, determine that view data is to catch image with the threshold value of being equal to or less than.
The view data definite process flow diagram handled of Figure 14 for carrying out by image processing apparatus 1300 by the regular line zone of using view data.
Rule line determining unit 1302 is determined regular line zone (step S1403) from binary picture data.Characteristic feature acquiring unit 1303 initiation parameter n are zero (step S1404).Characteristic feature acquiring unit 1303 is and the benchmark of image is caught in initialization and the benchmark of real electronic image is zero.The benchmark that image is caught in proposition is to be used for determining that image is to catch image, and the benchmark that proposes real electronic image is to be used for determining that image is real electronic image.
Characteristic feature acquiring unit 1303 determines that parameter n are whether greater than the number (step S1405) of regular line.When the number of regular line is equal to or less than parameter n (step S1405 denys), 1303 scannings of characteristic feature acquiring unit are corresponding to the image-region (step S1406) of the input image data in the regular line zone of binary picture data.
Subsequently, characteristic feature acquiring unit 1303 calculates the distribution value (step S1407) of the pixel value in scanning area.
Whether the distribution value that determining unit 1304 is determined to calculate is greater than predetermined threshold (step S1408).When distribution value during greater than predetermined threshold value (step S1408 is), determining unit 1304 increases the benchmark (step S1409) of catching image.On the other hand, when the distribution value is equal to or less than predetermined threshold value (step S1408 denys), determining unit 1304 increases the benchmark (step S1410) of real electronic image.
Characteristic feature acquiring unit 1303 increases parameter n (step S1411), and whether definite parameter n is greater than the number (step S1405) of regular line.
As parameter n during greater than the number of regular line (step S1405 is), whether the benchmark that determining unit 1304 determines to catch image is greater than the benchmark (step S1412) of real electronic image.When the benchmark of catching image during greater than the benchmark of real electronic image (step S1412 is), determining unit 1304 is determined images for catching image (step S1413), and the end process process.
On the other hand, when the benchmark of catching image is equal to or less than the benchmark of real electronic image (step S1412 denys), determining unit 1304 determines that images are real electronic image (step S1414), and the end process process.
According to the 3rd embodiment, except obtaining the effect of in first and second embodiment, describing, but speed up processing also, and reduce storer and use.Especially, image processing apparatus 1300 can determine exactly the input view data catch image or real electronic image.Correspondingly, can select appropriate image processing according to the characteristic feature of input picture.Further, even when the view data of input when being binary picture, image processing apparatus 1300 can determine that also the view data of importing catches image or real electronic image.
Having illustrated in the 3rd embodiment based on the pixel value in regular line zone distributes the view data of definite input to catch image or real electronic image.Yet, can also use the further feature in the regular line zone except that pixel value distributes.Next the width of the rule-based line of explanation is determined a kind of method of the type of view data in the fourth embodiment of the present invention.
Figure 15 is the functional-block diagram according to the image processing apparatus 1500 of the 4th embodiment.Image processing apparatus 1500 and image processing apparatus 1300 different being, image processing apparatus 1500 comprises characteristic feature acquiring unit 1502 and determining unit 1503, and wherein the operation of characteristic feature acquiring unit 1502 and determining unit 1503 is different with determining unit 1304 with characteristic feature acquiring unit 1303.Identical parts are endowed same reference marker, and have omitted the explanation for them.
Suppose image processing apparatus 1500 processing binary picture datas.
In the method for in the 3rd embodiment, describing,, therefore come the view data of definite input to catch image or real electronic image based on color (brightness) because the view data of input is the multi-level images data.Yet if the view data of input is a binary picture data, said method is just inapplicable.According to the 4th embodiment, even when the view data of input when being binary picture data, image processing apparatus 1500 can determine that also the view data of importing catches image or real electronic image.
Characteristic feature acquiring unit 1502 obtains the variable quantity (distribution value) of the width of regular line from the regular line zone of binary picture data.Especially, the variable quantity of characteristic feature acquiring unit 1502 width of computation rule line from the regular line zone of determining by regular line zone determining unit 1302.In other words, characteristic feature acquiring unit 1502 is measured the width of every regular line rather than the length of every regular line.
Figure 16 is the example of the binary picture data obtained by image processing apparatus 1500.Regular line in regular line zone 1601 as shown in figure 16 is processed to obtain the variable quantity of width.
Figure 17 is the enlarged diagram of the part rule line in regular line zone 1601.If regular line is horizontally disposed, then can detect the width of regular line by the spanning length of measuring the regular line that passes through in the horizontal direction.In example shown in Figure 17, the width that characteristic feature acquiring unit 1502 is measured corresponding to the regular line of each predetermined space indicated by the arrow.As can be seen from Figure 17 differ from one another with width 1702 as the width of measuring as a result 1701.
Characteristic feature acquiring unit 1502 is measured the width of the regular line that needs measurement.If regular line vertically is provided with, then obtain the width of regular line by the spanning length of measuring regular line in the horizontal direction.If regular line tilts, the horizontal spanning length of vertically disposed regular line and the width of the regular line on the technical meaning are inequality.
The measurement width of determining unit 1503 rule-based lines (the distribution value of the point on the regular line) determines that image catches image or real electronic image.Especially, pre-determine threshold value, and when the variable quantity of regular line width equaled predetermined threshold or be wideer than threshold value, determining unit 1503 determined that images are to catch image.In this way, do not need the absolute value of regular line width, and only need the variable quantity of width.Even when regular line slight inclination, can obtain the variable quantity of width therefore.Be identified for determining the threshold value of variable quantity based on experimental result.
Figure 18 is for coming the carries out image data to determine the process flow diagram of handling by image processing apparatus 1500 by the regular line zone of using view data.
Rule line zone determining unit 1302 is determined regular line zone (step S1802) from binary picture data.Characteristic feature acquiring unit 1502 initiation parameter n are zero (step S1803).Characteristic feature acquiring unit 1502 is and the benchmark of image is caught in initialization and the benchmark of real electronic image is zero.The benchmark that image is caught in proposition is to be used for determining that image is to catch image, and the benchmark that proposes real electronic image is to be used for determining that image is real electronic image.
Characteristic feature acquiring unit 1502 determines that parameter n are whether greater than the number (step S1804) of regular line.When parameter n is equal to or less than the number of regular line (step S1804 denys), the width (step S1805) of regular line on the precalculated position in the regular line zone of characteristic feature acquiring unit 1502 measurements binary picture data in the vertical.
Characteristic feature acquiring unit 1502 calculates at the variable quantity of measuring on the width (step S1806).
Whether the variable quantity that determining unit 1503 is determined to calculate is greater than predetermined threshold (step S1807).When variable quantity during greater than predetermined threshold value (step S1807 is), determining unit 1503 increases the benchmark (step S1808) of catching image.On the other hand, when variable quantity is equal to or less than predetermined threshold value (step S1807 denys), determining unit 1503 increases the benchmark (step S1809) of real electronic image.
Characteristic feature acquiring unit 1502 increases parameter n (step S1810), and whether definite parameter n is greater than the number (step S1804) of regular line.
As parameter n during greater than the number of regular line (step S1804 is), whether the benchmark that determining unit 1503 determines to catch image is greater than the benchmark (step S1811) of real electronic image.When the benchmark of catching image during greater than the benchmark of real electronic image (step S1811 is), determining unit 1503 is determined images for catching image (step S1812), and terminal procedure.
On the other hand, when the benchmark of catching image is equal to or less than the benchmark of real electronic image (step S1811 denys), determining unit 1503 determines that images are real electronic image (step S1813), and terminal procedure.
According to the 4th embodiment, even when view data is binary picture data, image processing apparatus 1500 can determine exactly that also image catches image or real electronic image.
If the view data of input is a binary picture data, then graphics processing unit 205 need not to carry out gray scale processing or scale-of-two processing.Yet graphics processing unit 205 can be carried out suitable crooked correction or other Flame Image Process to catching image.
In the 4th embodiment, illustrated based on the variable quantity of the width of the regular line in the binary picture data and determined that image catches image or real electronic image.Yet, can carry out based on the variable quantity of the width of the regular line in the multi-level images data except that binary picture data and above-mentionedly determine.
Figure 19 is the functional-block diagram of image processing apparatus 1900 according to a fifth embodiment of the invention.Image processing apparatus 1900 also comprises binary unit 1901 except the setting of image processing apparatus 1500.Identical parts are endowed same reference marker, and have omitted the explanation about them.
If the view data of being obtained by image acquisition unit 201 is the multi-level images data, then 1901 pairs of multi-level images data of binary unit are carried out the scale-of-two processing.
Other processing is identical with the processing of describing in the 4th embodiment, and no longer repeats identical explanation.According to the 5th embodiment, even when being the multi-level images data, view data can determine also that view data catches image or real electronic image.
Further, when the multi-level images data are carried out binarization, can be worth to determine the type of view data based on the distribution in the presumptive area of the binarization view data except that the wide variety amount in regular line zone.Especially, also can be worth to carry out and above-mentionedly determine based on the distribution of the character color in binary picture data.
In the 3rd to the 5th embodiment, described by determining regular line zone and whole regular lines being carried out above-mentioned processing determine that view data catches image or real electronic image.Yet, also can use other processing procedure.
First modified example of the 3rd to the 5th embodiment is described below.Determine the priority order of regular line to be processed according to the image processing apparatus of first modified example, distribution value or variable quantity according to the priority order molded breadth, determine view data whether for catching image based on result calculated, and after image is confirmed as catching image termination procedure immediately.Substantially the same according to the image processing apparatus of first modified example with the configuration of describing with reference to Figure 13, Figure 15 and Figure 19.
Characteristic feature acquiring unit according to the image processing apparatus of first modified example comes the ordering rule line according to the descending order of the pixel (or zone) in the regular line zone of being determined by regular line zone determining unit.Subsequently, determining unit is come the variable quantity of the width of the pixel distribution value of computation rule line or regular line according to the descending order of pixel (zone), thereby determines whether distribution value or variable quantity surpass predetermined threshold value.When distribution value or variable quantity surpassed predetermined threshold value, determining unit determined that view data is to catch image, and need not the Else Rule line carried out to handle to get final product terminal procedure.
Regular line with big zone is considered to wide and long regular line, thereby can think that the definite result from above-mentioned wide and long regular line is more reliable than the definite result from the regular line with zonule.
When determining unit is finished processing to whole regular lines according to priority order, and definite view data is not when catching image, and determining unit determines that view data is real electronic image.
As mentioned above, obtain higher priority by making the result that obtains from the regular line with big zone than the result that obtains from the regular line with zonule, determining unit can be carried out deterministic process to regular line.
The cause of obtaining above-mentioned priority is described below.Suppose that a regular line has the width of a point, and owing to the slight inclination of regular line makes its width become two points.Further, the regular line of supposing other has the width of 20 points and becomes 21 points owing to slight inclination makes it.In this case, the distribution value of the whole width of preceding a kind of regular line is a kind of bigger than the back, determines that view data is to catch image with leading to errors.Obtain higher priority by making, can prevent above-mentioned mistake from definite result of regular line with big zone.Especially, if determine the priority order of regular line based on the pixel value in regular line (being the zone of rule-based line), even when the width of regular line because slight inclination and when first becomes two points, then the priority order step-down of above-mentioned regular line.Therefore, even when having the regular line slight inclination of zonule can not determine that view data is to catch image yet.As a result, the situation that can prevent from mistakenly view data to be defined as based on the regular line of slight inclination to catch image occurs.
Further, when pre-determined number detect expression when catching the characteristic feature of image, but not determine and termination procedure when controlling when carrying out based on a regular line, can determine that view data is to catch image.
In addition, the characteristic feature acquiring unit can be by getting rid of some points but not on the scanning rule line all points calculate distribution value or variable quantity.As a result, but speed up processing.
According to first embodiment, when determining that based on regular line view data is when catching image, the image processing apparatus termination procedure with high priority.Therefore, the processing time can be reduced.
Further, store regular line to be processed, thereby can keep processing speed efficiently and handle degree of accuracy.
Also can come the ordering rule line according to the other factors except their zone.Second modified example of the 3rd to the 5th embodiment is described below.Image processing apparatus according to second modified example comes its ordering according to the length of regular line.
Characteristic feature acquiring unit according to second modified example comes the ordering rule line according to the descending order of the length of regular line in the regular line zone of being determined by the regular line zone determining unit of image processing apparatus.Image processing apparatus really order unit from calipers then the order that begins of line come the pixel value of computation rule line to distribute or the variable quantity of regular line width.Subsequently, determining unit determines whether distribution value or variable quantity surpass predetermined threshold value.When distribution value or variable quantity surpassed predetermined threshold, determining unit determined that view data is to catch image, and need not other regular line carried out to handle to get final product termination.
Some of pixel that also can be by getting rid of view data in the vertical in second modified example put and calculated distribution value or variable quantity.
According to second modified example, can be used for determining that whether view data is the variable quantity of catching computation rule line width on one section regular line of image.Further, can improve the degree of accuracy and the minimizing processing time of deterministic process.
The modified example of the 3rd to the 5th embodiment is described below.Image processing apparatus according to the 3rd modified example need not the ordering rule line and the normalized number certificate.
According to the characteristic feature acquiring unit of the image processing apparatus of the 3rd modified example not to the regular line ordering in the regular line zone of determining by regular line zone determining unit, no matter the length of regular line and the pixel value that calculates regular line from 20 parts of whole regular line distributes or the variable quantity of regular line width.Subsequently, image processing apparatus really order unit determine whether distribution value and variable quantity surpass predetermined threshold value.The part number that is used to carry out deterministic process is not limited to 20, and can use any predetermined number.
According to the 3rd modified example, even when the length of regular line changed, the data volume that obtain can remain constant.Therefore, no matter the length of regular line can stably be carried out definite.
The 4th modified example of the 3rd to the 5th embodiment is described below.With in the 3rd to the 5th embodiment and first to the 3rd modified example, describe different, whole but a limited number of regular line is not carried out according to the image processing apparatus of the 4th modified example and to be determined to handle.
Especially, when the view data of input was form or the document as the classification paper, the view data of input comprised many regular lines.
If the number of regular line increases, then data volume also increases thereupon, has improved the degree of accuracy of definite processing.Yet,, need the more processing time and increased to handle load if extract whole regular lines and all handle the regular line that extracts.Especially, if handle tens of regular lines, also be disadvantageous even then improve the processing degree of accuracy and increase the processing load with same widths and same intervals.
The characteristic feature acquiring unit of image processing apparatus is selected regular line to be processed from the regular line zone of being determined by the regular line zone determining unit of image processing apparatus, and the regular line of choosing is carried out processing.Especially, can get rid of the mode of short regular line and come the selective rule line, or when regular line parallel is provided with, handle the regular line of every predetermined number.In the above described manner, only the regular line of necessity is carried out processing according to the image processing apparatus of the 4th modified example.
For example, can determine that regular line that the detection to predetermined ratio obtains is carried out to handle, promptly when detecting 100 regular lines, only handle 20 regular lines.If determine that from 20 regular lines ten or more regular lines are as catching image, then terminal procedure.If even definite view data is not to catch image after handling 20 regular lines, need not remaining 80 regular lines processing are got final product terminal procedure.
Can be single ground or combination with one another get up to use together the foregoing description and modified example.
For example, when detecting 100 regular lines, can carry out definite the processing to 20 regular lines according to the priority order of their length.
According to the 4th modified example, can keep processing speed effectively and handle degree of accuracy by giving up some regular lines.
According to an aspect of the present invention, can carry out appropriate image processing according to the type of the view data of importing.As a result, can reduce processing time and raising processing degree of accuracy.
Although the present invention is by clearly having described its disclosure with reference to specific embodiment is complete, but appended claim is not limited to this, but is used for explaining all modified examples that can draw for those skilled in the art and the optional structure that falls into the basis instruction that proposes in the present invention.
Graphics processing unit
Claims (17)
1. image processing apparatus comprises:
Image acquisition unit is used to obtain view data;
The feature acquiring unit is used to obtain the characteristic feature of view data;
Determining unit is used for determining based on characteristic feature the type of view data; And
Graphics processing unit is used for type according to view data to the view data carries out image processing.
2. image processing apparatus according to claim 1 further comprises:
The zone determining unit, be used for determining view data extraction the target area of characteristic feature, wherein
The feature acquiring unit obtains characteristic feature from this target area.
3. image processing apparatus according to claim 2, wherein
The feature acquiring unit obtains the distribution value of distribution of remarked pixel value as characteristic feature, and
Whether determining unit surpasses the type that threshold value is determined view data based on the distribution value.
4. according to claim 2 or 3 described image processing apparatus, wherein
The zone determining unit determines to be included in zone in the background area of view data as the target area.
5. image processing apparatus according to claim 1 further comprises:
The color cluster unit is used for view data is carried out color cluster, and generates the color cluster that is illustrated in the color classification in the view data, wherein
The feature acquiring unit obtains characteristic feature for each color cluster.
6. image processing apparatus according to claim 5, wherein
The feature acquiring unit obtains the distribution value of distribution of remarked pixel value as characteristic feature, and
Whether determining unit surpasses the type that threshold value is determined view data based on the distribution value.
7. image processing apparatus according to claim 1 further comprises:
Binary unit is used for that view data is carried out scale-of-two and handles, and generate binary picture data from view data; And
Rule line zone determining unit is used for determining the regular line zone in binary picture data, wherein
The feature acquiring unit obtains characteristic feature from the zone corresponding to the view data in regular line zone.
8. image processing apparatus according to claim 1 further comprises:
Rule line zone determining unit is used for when view data is binary picture data, determines the regular line zone in binary picture data, wherein
The feature acquiring unit obtains the variable quantity of width of the regular line on the longitudinal direction in regular line zone as characteristic feature.
9. image processing apparatus according to claim 1 further comprises:
Binary unit is used for when view data is the multi-level images data binarization view data to generate binary picture data; And
Rule line zone determining unit is used for determining the regular line zone in binary picture data, wherein
The feature acquiring unit obtains the variable quantity of width of the regular line on the longitudinal direction in regular line zone as characteristic feature.
10. according to each the described image processing apparatus in the claim 7 to 9, wherein
The feature acquiring unit obtains the distribution value of distribution of remarked pixel value as characteristic feature, and
Whether determining unit surpasses the type that threshold value is determined view data based on the distribution value.
11. according to each the described image processing apparatus in the claim 7 to 10, wherein,
When regular line zone determining unit was determined a plurality of regular line zone in binary picture data, the feature acquiring unit was from obtaining characteristic feature based on one of them of predetermined benchmark predetermined rule line zone.
12. according to each the described image processing apparatus in the claim 1 to 11, wherein
Graphics processing unit according to the type of view data control gray scale handle, scale-of-two is handled and crooked correction is handled at least one of them.
13. an image processing method comprises step:
Receive view data;
Obtain the characteristic feature of view data;
Determine the type of view data based on characteristic feature; And
According to the type of view data to the view data carries out image processing.
14. image processing method according to claim 13 further comprises
Determine in the view data extraction the target area of characteristic feature, wherein
Above-mentioned obtaining step comprises obtain characteristic feature from the target area.
15. image processing method according to claim 13 further comprises the following steps:
The binarization view data is to generate binary picture data from view data; And
Determine the regular line zone in the binary picture data, wherein
Above-mentioned obtaining step comprises from the zone corresponding to the view data in regular line zone and obtains characteristic feature.
16. image processing method according to claim 13 further comprises the following steps:
The binarization view data is to generate binary picture data when view data is the multi-level images data; And
Determine the regular line zone in the binary picture data, wherein
Above-mentioned obtaining step comprise obtain in regular line zone vertically on the variable quantity of width of regular line as characteristic feature.
17. a computer program comprises the computer usable medium with the computer readable program code that embodies in medium, when carrying out this computer readable program code, computing machine is carried out:
Receive view data;
Obtain the characteristic feature of view data;
Determine the type of view data based on characteristic feature; And
According to the type of view data to the view data carries out image processing.
Applications Claiming Priority (6)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2007053978 | 2007-03-05 | ||
| JP2007053978 | 2007-03-05 | ||
| JP2007-053978 | 2007-03-05 | ||
| JP2007325145A JP2008252862A (en) | 2007-03-05 | 2007-12-17 | Image processing apparatus, image processing method, and image processing program |
| JP2007-325145 | 2007-12-17 | ||
| JP2007325145 | 2007-12-17 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101261682A true CN101261682A (en) | 2008-09-10 |
| CN101261682B CN101261682B (en) | 2012-03-28 |
Family
ID=39962131
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008100825564A Expired - Fee Related CN101261682B (en) | 2007-03-05 | 2008-03-03 | Image processing apparatus, image processing method |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP2008252862A (en) |
| CN (1) | CN101261682B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101976348A (en) * | 2010-10-21 | 2011-02-16 | 中国科学院深圳先进技术研究院 | Image clustering method and system |
| CN102204271A (en) * | 2011-06-28 | 2011-09-28 | 华为终端有限公司 | A method for controlling user equipment and a device thereof |
| CN104778473A (en) * | 2014-01-09 | 2015-07-15 | 深圳市中瀛鑫科技股份有限公司 | Image binarization method, device and video analysis system |
| CN104915675A (en) * | 2014-03-14 | 2015-09-16 | 欧姆龙株式会社 | Image processing device, image processing method, and image processing program |
| CN105260709A (en) * | 2015-09-28 | 2016-01-20 | 北京石油化工学院 | Water meter detecting method, apparatus, and system based on image processing |
| CN105260710A (en) * | 2015-09-28 | 2016-01-20 | 北京石油化工学院 | Water meter detecting method, apparatus, and system based on image processing |
| CN110602358A (en) * | 2019-08-28 | 2019-12-20 | 维沃移动通信有限公司 | Image acquisition method and electronic equipment |
| CN113270169A (en) * | 2021-05-26 | 2021-08-17 | 青岛海信医疗设备股份有限公司 | Image display method, device, processing equipment and medium |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5867045B2 (en) * | 2011-12-12 | 2016-02-24 | 富士ゼロックス株式会社 | Image processing apparatus and program |
| CN103778436B (en) * | 2014-01-20 | 2017-04-05 | 电子科技大学 | A kind of pedestrian's attitude detecting method based on image procossing |
Family Cites Families (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3164744B2 (en) * | 1994-12-21 | 2001-05-08 | 株式会社東芝 | Image forming device |
| JP3504054B2 (en) * | 1995-07-17 | 2004-03-08 | 株式会社東芝 | Document processing apparatus and document processing method |
| JPH09326922A (en) * | 1996-06-03 | 1997-12-16 | Fuji Xerox Co Ltd | Image processor |
| JP2000148980A (en) * | 1998-11-12 | 2000-05-30 | Fuji Photo Film Co Ltd | Image processing method, image processor and recording medium |
| JP2000261653A (en) * | 1999-03-09 | 2000-09-22 | Hitachi Ltd | Image processing unit |
| JP3918381B2 (en) * | 1999-10-25 | 2007-05-23 | 富士ゼロックス株式会社 | Image processing apparatus and image processing method |
| JP4001446B2 (en) * | 2000-02-09 | 2007-10-31 | 株式会社リコー | Method, apparatus and computer-readable recording medium for specifying image background color |
| JP2001251507A (en) * | 2000-03-03 | 2001-09-14 | Fujitsu Ltd | Image processor |
| JP3904840B2 (en) * | 2000-08-15 | 2007-04-11 | 富士通株式会社 | Ruled line extraction device for extracting ruled lines from multi-valued images |
| JP2003060980A (en) * | 2001-08-21 | 2003-02-28 | Olympus Optical Co Ltd | Image processing system |
| CN1276382C (en) * | 2001-10-11 | 2006-09-20 | 皇家飞利浦电子股份有限公司 | Method and apparatus for discriminating between different regions of an image |
| JP3865215B2 (en) * | 2001-12-14 | 2007-01-10 | 株式会社リコー | Image distortion correction apparatus, program, storage medium, and image distortion correction method |
| JP2003219191A (en) * | 2002-01-21 | 2003-07-31 | Sharp Corp | Image processor and image forming apparatus comprising it |
| JP4208520B2 (en) * | 2002-08-26 | 2009-01-14 | キヤノン株式会社 | Image processing apparatus, image processing method, program, and storage medium |
| JP2004129022A (en) * | 2002-10-04 | 2004-04-22 | Pfu Ltd | Image reading device |
| JP4265237B2 (en) * | 2003-02-27 | 2009-05-20 | ソニー株式会社 | Image processing apparatus and method, learning apparatus and method, recording medium, and program |
| CN1286064C (en) * | 2003-07-23 | 2006-11-22 | 西北工业大学 | An image retrieval method based on marked interest point |
| JP2006072418A (en) * | 2004-08-31 | 2006-03-16 | Make Softwear:Kk | Image composition method and device |
| JP4651407B2 (en) * | 2005-02-21 | 2011-03-16 | 株式会社リコー | Image processing apparatus, computer program, and storage medium |
| JP4849818B2 (en) * | 2005-04-14 | 2012-01-11 | イーストマン コダック カンパニー | White balance adjustment device and color identification device |
| JP4730525B2 (en) * | 2005-06-13 | 2011-07-20 | 富士ゼロックス株式会社 | Image processing apparatus and program thereof |
| JP5111794B2 (en) * | 2005-08-08 | 2013-01-09 | 株式会社東芝 | Paper sheet identification device, paper sheet identification method, and dictionary creation method |
-
2007
- 2007-12-17 JP JP2007325145A patent/JP2008252862A/en active Pending
-
2008
- 2008-03-03 CN CN2008100825564A patent/CN101261682B/en not_active Expired - Fee Related
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101976348A (en) * | 2010-10-21 | 2011-02-16 | 中国科学院深圳先进技术研究院 | Image clustering method and system |
| CN102204271A (en) * | 2011-06-28 | 2011-09-28 | 华为终端有限公司 | A method for controlling user equipment and a device thereof |
| CN104778473A (en) * | 2014-01-09 | 2015-07-15 | 深圳市中瀛鑫科技股份有限公司 | Image binarization method, device and video analysis system |
| CN104915675A (en) * | 2014-03-14 | 2015-09-16 | 欧姆龙株式会社 | Image processing device, image processing method, and image processing program |
| CN105260709A (en) * | 2015-09-28 | 2016-01-20 | 北京石油化工学院 | Water meter detecting method, apparatus, and system based on image processing |
| CN105260710A (en) * | 2015-09-28 | 2016-01-20 | 北京石油化工学院 | Water meter detecting method, apparatus, and system based on image processing |
| CN105260710B (en) * | 2015-09-28 | 2018-08-28 | 北京石油化工学院 | Water meter calibration method, apparatus based on image procossing and system |
| CN105260709B (en) * | 2015-09-28 | 2018-11-02 | 北京石油化工学院 | Water meter calibration method, apparatus based on image procossing and system |
| CN110602358A (en) * | 2019-08-28 | 2019-12-20 | 维沃移动通信有限公司 | Image acquisition method and electronic equipment |
| CN110602358B (en) * | 2019-08-28 | 2021-06-04 | 维沃移动通信有限公司 | Image acquisition method and electronic equipment |
| CN113270169A (en) * | 2021-05-26 | 2021-08-17 | 青岛海信医疗设备股份有限公司 | Image display method, device, processing equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2008252862A (en) | 2008-10-16 |
| CN101261682B (en) | 2012-03-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101261682B (en) | Image processing apparatus, image processing method | |
| JP6139396B2 (en) | Method and program for compressing binary image representing document | |
| JP5934762B2 (en) | Document modification detection method by character comparison using character shape characteristics, computer program, recording medium, and information processing apparatus | |
| JP5701181B2 (en) | Image processing apparatus, image processing method, and computer program | |
| US8675256B2 (en) | Image processing apparatus, image processing method, and storage medium | |
| US20090252381A1 (en) | Paper-sheet stain detecting apparatus and method | |
| CN101551859A (en) | Image recognition device and image retrieval device | |
| CN110598566A (en) | Image processing method, device, terminal and computer readable storage medium | |
| CN100530219C (en) | Image processing apparatus | |
| US20080310685A1 (en) | Methods and Systems for Refining Text Segmentation Results | |
| US8229214B2 (en) | Image processing apparatus and image processing method | |
| US8830529B2 (en) | Image processing device for accurately identifying region in image without increase in memory requirement | |
| US8705134B2 (en) | Method of processing an image to clarify text in the image | |
| JP5887242B2 (en) | Image processing apparatus, image processing method, and program | |
| JP5788367B2 (en) | Long object number measuring device, long object number measuring method, and computer program | |
| US9277074B2 (en) | Image processing apparatus, method, and medium determining whether image data of a page to be processed is blank and contains a foreground object and transmitting the foreground object obtained by removing a background object | |
| CN111209912A (en) | Method for removing long interference lines of Chinese character and picture | |
| US11800036B2 (en) | Determining minimum scanning resolution | |
| CN106934814B (en) | Background information identification method and device based on image | |
| JP2005150837A (en) | Document processor | |
| CN111340677A (en) | Video watermark detection method and device, electronic equipment and computer readable medium | |
| KR100850173B1 (en) | The method and apparatus for image binarization | |
| JP2010218106A (en) | Image processing apparatus, image processing method and program | |
| JP2613211B2 (en) | Image input device | |
| CN115861270A (en) | Image detection method and device based on spatial distribution and storage medium thereof |
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: 20120328 Termination date: 20170303 |