CN104751165A - Back-through detection method and device - Google Patents
Back-through detection method and device Download PDFInfo
- Publication number
- CN104751165A CN104751165A CN201310744057.8A CN201310744057A CN104751165A CN 104751165 A CN104751165 A CN 104751165A CN 201310744057 A CN201310744057 A CN 201310744057A CN 104751165 A CN104751165 A CN 104751165A
- Authority
- CN
- China
- Prior art keywords
- pixel
- point
- foreground
- direct picture
- background
- 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.)
- Pending
Links
Landscapes
- Character Input (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a back-through detection method and device. The back-through detection method and device comprises obtaining a front side image and a back side image which are aligned so as to obtain pixel pairs, wherein every pixel pair is formed by a first point and a corresponding second point; determining part of foreground pixels and background pixels in the front side image and the back side image; performing preset characteristic based modeling on the following pixels based on a determined result so as to form four models, wherein the following pixels comprise the first points and the second points which are foreground pixels, the first points and the second points which are background pixels, the first points which are foreground pixels and the second points which are background pixels and the first points which are background pixels and the second points which are foreground pixels; calculating the similarity of the pixel pairs which are not modeled with the four models according to the preset characteristics of the pixel pairs which are not modeled so as to determine the types of the pixel pairs which are not modeled; determining the second points which are corresponding to the first points determined to be the foreground pixels and are determined to be the background pixels to be back-though points on the back side image and determining the first points which are corresponding to the second points determined to be the foreground pixels and are determined to be the background pixels to be back-though points on the front side image.
Description
Technical field
Relate generally to image processing field of the present invention.Specifically, the present invention relates to the method and apparatus of the position of the saturating pixel of the back of the body of the dual-side image centering that a kind of detection is made up of direct picture and verso images.
Background technology
In the two-sided document with pros and cons, because document exists the saturating phenomenon of the back of the body, so the part or all of word etc. of another side can be seen from one side.The readability that the saturating phenomenon of this back of the body can affect document and the automatic character recognition that document is carried out.Therefore, need to process the saturating pixel of the back of the body wherein, to improve the readability of document or to improve the accuracy of automatic character recognition.
The prerequisite processed the saturating pixel of the back of the body accurate location carrying on the back saturating pixel detected, namely judges which pixel is the saturating pixel of the back of the body.
Traditional method mainly chooses suitable threshold value, by the pixel value of pixel and threshold value being made comparisons, distinguishing word and carrying on the back thoroughly.This method is obviously more coarse.Because be that difference between the saturating pixel of the back of the body in the foreground pixel of representative and the saturating situation of the serious back of the body is less with word, and difference between the saturating pixel of the back of the body in background pixel and the saturating situation of the slight back of the body is also not obvious.Therefore, the treatment effect of traditional method is not good.
Expect a kind of can with detect compared with high-accuracy the position of the saturating pixel of the back of the body of the dual-side image centering be made up of direct picture and verso images the saturating detection method of the back of the body and the back of the body saturating checkout equipment.
Summary of the invention
Give hereinafter about brief overview of the present invention, to provide about the basic comprehension in some of the present invention.Should be appreciated that this general introduction is not summarize about exhaustive of the present invention.It is not that intention determines key of the present invention or pith, and nor is it intended to limit the scope of the present invention.Its object is only provide some concept in simplified form, in this, as the preorder in greater detail discussed after a while.
The object of the invention is the problems referred to above for prior art, propose and a kind ofly accurately can detect the saturating detection method of the back of the body of carrying on the back saturating location of pixels and the saturating checkout equipment of the back of the body.
To achieve these goals, according to an aspect of the present invention, provide the saturating detection method of a kind of back of the body, the saturating detection method of this back of the body comprises: the direct picture and the verso images that obtain the dual-side image centering of having alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel pair formed with corresponding described second point; Determine the part foreground pixel in described direct picture and described verso images and part background pixel; Based on above-mentioned determination result, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel; For the pixel pair be not modeled, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models, respectively to determine the type that this pixel is right; The back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture is saturating.
According to another aspect of the present invention, provide the saturating checkout equipment of a kind of back of the body, the saturating checkout equipment of this back of the body comprises: acquisition device, be configured to: the direct picture and the verso images that obtain the dual-side image centering of having alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel pair formed with corresponding described second point; Partial pixel determining device, is configured to: determine the part foreground pixel in described direct picture and described verso images and part background pixel; Model building device, be configured to: based on the determination result of described partial pixel determining device, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel; Type determination device, is configured to: for the pixel pair be not modeled, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models respectively, to determine the type that this pixel is right; Decision maker, be configured to: the back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture is saturating.
In addition, according to a further aspect in the invention, a kind of storage medium is additionally provided.Described storage medium comprises machine-readable program code, and when performing described program code on messaging device, described program code makes described messaging device perform according to said method of the present invention.
In addition, in accordance with a further aspect of the present invention, a kind of program product is additionally provided.Described program product comprises the executable instruction of machine, and when performing described instruction on messaging device, described instruction makes described messaging device perform according to said method of the present invention.
Accompanying drawing explanation
Below with reference to the accompanying drawings illustrate embodiments of the invention, above and other objects, features and advantages of the present invention can be understood more easily.Parts in accompanying drawing are just in order to illustrate principle of the present invention.In the accompanying drawings, same or similar technical characteristic or parts will adopt same or similar Reference numeral to represent.In accompanying drawing:
Fig. 1 shows the process flow diagram according to the saturating detection method of the back of the body of the first embodiment of the present invention;
Fig. 2 shows the process flow diagram of the saturating detection method of the back of the body according to a second embodiment of the present invention;
Fig. 3 shows the block diagram of the saturating checkout equipment of the back of the body according to a first embodiment of the present invention;
Fig. 4 shows the block diagram of the saturating checkout equipment of the back of the body according to a second embodiment of the present invention; And
Fig. 5 shows and can be used for implementing the schematic block diagram according to the computing machine of the method and apparatus of the embodiment of the present invention.
Embodiment
To be described in detail one exemplary embodiment of the present invention by reference to the accompanying drawings hereinafter.For clarity and conciseness, all features of actual embodiment are not described in the description.But, should understand, must make a lot specific to the decision of embodiment in the process of any this actual embodiment of exploitation, to realize the objectives of developer, such as, meet those restrictive conditions relevant to system and business, and these restrictive conditions may change to some extent along with the difference of embodiment.In addition, although will also be appreciated that development is likely very complicated and time-consuming, concerning the those skilled in the art having benefited from present disclosure, this development is only routine task.
At this, also it should be noted is that, in order to avoid the present invention fuzzy because of unnecessary details, illustrate only in the accompanying drawings with according to the closely-related apparatus structure of the solution of the present invention and/or treatment step, and eliminate other details little with relation of the present invention.In addition, also it is pointed out that the element described in an accompanying drawing of the present invention or a kind of embodiment and feature can combine with the element shown in one or more other accompanying drawing or embodiment and feature.
Basic thought of the present invention first determines those correct foreground pixel and background pixels certainly, and according to these foreground pixels and background pixel modeling, the model constructed by utilization judges the type of remaining pixel.In the process, the feature of carrying on the back saturating pixel can be utilized, and reduce calculated amount by preferred process object and improve processing speed.
Below with reference to Fig. 1, the flow process according to the saturating detection method of the back of the body of the first embodiment of the present invention is described.
Fig. 1 shows the process flow diagram according to the saturating detection method of the back of the body of the first embodiment of the present invention.As shown in Figure 1, according to the saturating detection method of the back of the body of the first embodiment of the present invention, comprise the steps: to obtain the direct picture and verso images that have alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel formed with corresponding described second point (step S1); Determine the part foreground pixel in described direct picture and described verso images and part background pixel (step S2); Based on the determination result in step S2, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel (step S3); For the pixel pair be not modeled, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models, respectively with the type (step S4) determining that this pixel is right; The back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture thoroughly (step S5).
In step sl, obtain the direct picture and verso images that have alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel pair formed with corresponding described second point.
As mentioned above, the present invention to be processed to as if there is the two-sided document of pros and cons.Therefore, input is the dual-side image pair be made up of direct picture and verso images, direct picture and the verso images back side each other.
In direct picture, the contents such as word are foreground pixels, and other pixel is background pixel.Similarly, in verso images, the contents such as word are foreground pixels, and other pixel is background pixel.If the foreground pixel back of the body of verso images is thoroughly in direct picture, and does not overlap with the foreground pixel of direct picture, then such pixel is that the back of the body of foreground pixel on direct picture of verso images is saturating.If the foreground pixel back of the body of direct picture is thoroughly in verso images, and does not overlap with the foreground pixel of verso images, then such pixel is that the back of the body of foreground pixel on verso images of direct picture is saturating.
It should be noted that if the foreground pixel back of the body of verso images is thoroughly in direct picture, and overlap with the foreground pixel of direct picture, then such pixel is the foreground pixel of direct picture, is not that the back of the body on direct picture is saturating.Similarly, if the foreground pixel back of the body of direct picture is thoroughly in verso images, and overlap with the foreground pixel of verso images, then such pixel is the foreground pixel of verso images, is not that the back of the body on verso images is saturating.
Therefore, the saturating pixel of the back of the body on direct picture is determined by two points corresponding on pros and cons image: first on direct picture is background pixel, and second point corresponding with first on verso images is when being foreground pixel, this first on direct picture is that the foreground pixel back of the body on verso images is thoroughly to the saturating pixel of the back of the body on direct picture.
For simplified illustration, below, the point on direct picture is called the first point, the point on verso images is called second point.
Obviously, in order to judge that whether pixel is the saturating pixel of the back of the body, need first to understand the corresponding relation of the second point on direct picture first and verso images.
Therefore, in step sl, the global affine transformation method of such as Dubois and Pathak etc. can be adopted to align direct picture and verso images, make for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel pair formed with corresponding described second point.
Below, the pixel formed by the second point for first and correspondence is to processing.
In step s 2, the part foreground pixel in described direct picture and described verso images and part background pixel is determined.
As mentioned above, thought of the present invention first determines a part of sub pixel, then with reference to the feature of sub pixel, determines the type of other pixel.Therefore, in step s 2, part can determine that first the pixel of its type is extracted.
Step S2 has multiple implementation method.
For example, according to the magnitude relationship of each pixel value of first and the first predetermined prospect threshold value or the first background threshold, the type of pixel of described first can be determined.According to the magnitude relationship of the pixel value of each second point and the second predetermined prospect threshold value or the second background threshold, the type of pixel of described second point can be determined.
Specifically, if the pixel value of first is less than the first prospect threshold value, then determine first foreground pixel being direct picture.Further, if the pixel value of first is greater than the first background threshold, then first background pixel being direct picture is determined.
Similarly, if the pixel value of second point is less than the second prospect threshold value, then determine that second point is the foreground pixel of verso images.Further, if the pixel value of second point is greater than the second background threshold, then determine that second point is the background pixel of verso images.
Can, by the first and second prospect threshold values are arranged less, the first and second background threshold be arranged guarantee more greatly the accuracy that step S2 judges.
The size of adjustment first and second prospect threshold value, the first and second background threshold can also be passed through, carry out the quantity of the pixel extracted in set-up procedure S2, to meet modeling demand.
Those skilled in the art according to the needs of the knowledge of this area and actual conditions, should be able to design the first and second suitable prospect threshold values and the first and second background threshold.
The another kind of implementation method of step S2 is such as according to the magnitude relationship of each pixel value of first with the pixel value of corresponding second point, determines the part foreground pixel in described direct picture and verso images and part background pixel.
This is because the feature of carrying on the back pixel is that the foreground pixel that the saturating pixel of the back of the body on a face compares on another face producing it is shallow, namely pixel value is large.So, by comparing a pair and second points at first, pixel larger for the difference of and second point is defined as the foreground pixel carried on the back pixel and produce it at first.
Particularly, the business of the pixel value of described first and the pixel value of corresponding second point and the 3rd predetermined threshold value and the 4th threshold value can be made comparisons, described 3rd threshold value is less than 1, and described 4th threshold value is greater than 1.If described business is less than described 3rd threshold value, then described first is foreground pixel in described direct picture, and described second point is the background pixel in described verso images; If described business is greater than described 4th threshold value, then described second point is the foreground pixel in described verso images, and described first is background pixel in described direct picture.
For example, the 3rd threshold value can get 0.5, and the 4th threshold value can get 2.0.The present invention is not limited thereto.
Should be understood that the above-mentioned two kinds of modes in step S2 can be used with.Such as, Land use systems one carrys out Extraction parts background pixel, and Land use systems two carrys out Extraction parts foreground pixel.
Next, for the foreground pixel on the direct picture determined in step s 2 and verso images and background pixel, carry out classification model construction.
In step s3, based on the determination result in step S2, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel.
Here modeling method can adopt any suitable known method, does not do concrete restriction.As long as modeling method can go out the model of the right feature of sign four groups of pixels based on the feature construction that four groups of pixels are right, can be distinguished from each other and relatively judge that other pixels are more similar to which group with four groups of pixel centerings by feature.
Modeling based on feature can be the pixel value that pixel is right, or other suitable feature.
According to four groups of pixels to after constructing four models, model just can be utilized to judge the type that pixel to be determined is right.
In step s 4 which, for the pixel pair be not modeled, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models, respectively to determine the type that this pixel is right.
Specifically, the average pixel value that can calculate described four models first respectively to and variance; Then according to the average pixel value of described four models to variance, pixel value that this pixel is right, calculate this pixel respectively to the distance to described four models; Finally type right for this pixel is defined as the type that the pixel of model corresponding to minor increment is right.
Average pixel value is right to referring to that the mean value of all pixels of a model to the pixel value of each pixel (or second point) in (and second point form to) is formed at first at first.That is, the pixel value of all first in a model is averaged, and the pixel value of all second points is averaged, and the result be averaged forms average pixel value pair.
Distance can select the suitable distances such as mahalanobis distance.The present invention is not limited thereto.
Those skilled in the art can understand that minimum model of adjusting the distance to pixel is to the most similar model to pixel.Therefore, can determine that the type that type that this pixel is right is right with the pixel of this model is consistent.
By step S1-S4 above, to all pixels to classifying, each pixel is to belonging to one of following Four types: first and second point are foreground pixels, first and second point are background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel.
Like this, according to above-mentioned judged result, in step s 5, the pixel can selecting latter two type, to as comprising the pixel pair of carrying on the back saturating pixel, therefrom extracts the saturating pixel of the back of the body further.
Particularly, in step s 5, the back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture is saturating.
So far, by step S1-S5, detect the position of the saturating pixel of the back of the body of the dual-side image centering be made up of direct picture and verso images.
Here be also pointed out that the described direct picture that dual-side image centering comprises and described verso images can be gray level images, described pixel value is gray-scale value.
The described direct picture that described dual-side image centering comprises and described verso images also can be coloured images, and described pixel value is the value of each passage.
In addition, according to the judged result of step S1-S4, can be judged to be the prospect on described direct picture at first that is confirmed as foreground pixel, the second point being confirmed as foreground pixel is judged to be the prospect on described verso images.
That is, as secondary product of the present invention, the position of the foreground pixel on the described direct picture that described dual-side image centering comprises and described verso images also can be judged.
In the saturating detection method of the back of the body of the first embodiment of the present invention, in step s 4 which, for the pixel pair be not modeled, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models, respectively to determine the type that this pixel is right.
In the saturating detection method of the back of the body of the second embodiment of the present invention, the number of the handling object in step S4 can be reduced.This is because in a first embodiment, for all pixels be not modeled to processing.In fact, owing to the object of the invention is to detect the saturating pixel of the back of the body, and carry on the back pixel be all the foreground pixel back of the body through coming, as long as so detect all foreground pixels and judge that whether the point at the back side of its correspondence is the background pixel at the back side.
Fig. 2 shows the process flow diagram of the saturating detection method of the back of the body according to a second embodiment of the present invention.
As shown in Figure 2, the saturating detection method of the back of the body according to a second embodiment of the present invention, comprise the steps: to obtain the direct picture and verso images that have alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel formed with corresponding described second point (step S1); Determine the part foreground pixel in described direct picture and described verso images and part background pixel (step S2); Determine candidate's foreground pixel in described direct picture and described verso images (step S2 '); Based on the determination result in step S2, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel (step S3); For in described candidate's foreground pixel, pixel pair belonging to pixel except described part foreground pixel, the predetermined characteristic right according to this pixel, calculate the similarity of this pixel pair and described four models respectively, to determine type that this pixel is right (step S4 '); The back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture thoroughly (step S5).
Step S1 in the second embodiment of the present invention, S2, S3, S5 and the step S1 in the first embodiment of the present invention, S2, S3, S5 are identical, so adopt identical label and the descriptions thereof are omitted.
The difference of the second embodiment of the present invention and the first embodiment of the present invention is below described.
In the second embodiment of the present invention, upon step s 2, step S2 ' is added.
In step S2 ', determine the candidate's foreground pixel in described direct picture and described verso images.
Particularly, the first and second above-mentioned background threshold can be utilized, judge the candidate's foreground pixel in described direct picture and described verso images.
If the pixel value of described first is less than or equal to the first predetermined background threshold, then determine that be candidate's foreground pixel in described direct picture at described first; If the pixel value of described second point is less than or equal to the second predetermined background threshold, then determine that described second point is the candidate's foreground pixel in described verso images.
Like this, it may be all the pixel of foreground pixel that the candidate's foreground pixel judged in step S2 ' includes, and has been confirmed as the pixel of the part foreground pixel in described direct picture and described verso images before including in step s 2.
Therefore, be necessary in described candidate's foreground pixel, pixel belonging to pixel except described part foreground pixel is to judging again.
In step S4 ' (the step S4 corresponding to the first embodiment), for in described candidate's foreground pixel, pixel pair belonging to pixel except described part foreground pixel, the predetermined characteristic right according to this pixel, calculate the similarity of this pixel pair and described four models respectively, to determine the type that this pixel is right.
Like this, through step S1-S4 ', all foreground pixels can be extracted, and judge the type of the point at the back side that these foreground pixels are corresponding.In step s 5, just the saturating pixel of all back ofs the body can correspondingly be extracted.
Compared with the first embodiment, decrease certain calculated amount according to the saturating detection method of the back of the body of the second embodiment, but do not affect recall rate.
Below, with reference to Fig. 3 description saturating checkout equipment of the back of the body according to a first embodiment of the present invention.
Fig. 3 shows the block diagram of the saturating checkout equipment of the back of the body according to a first embodiment of the present invention.As shown in Figure 3, the saturating checkout equipment 300 of this back of the body of the saturating checkout equipment of the back of the body according to a first embodiment of the present invention 300 comprises: acquisition device 31, be configured to: the direct picture and the verso images that obtain the dual-side image centering of having alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel pair formed with corresponding described second point; Partial pixel determining device 32, is configured to: determine the part foreground pixel in described direct picture and described verso images and part background pixel; Model building device 33, be configured to: based on the determination result of described partial pixel determining device 32, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel; Type determination device 34, is configured to: for the pixel pair be not modeled, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models respectively, to determine the type that this pixel is right; Decision maker 35, be configured to: the back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture is saturating.
In one embodiment, partial pixel determining device 32 comprises threshold decision unit, this threshold decision unit is configured to: according to the magnitude relationship of each pixel value of first and the first predetermined prospect threshold value, determine the part foreground pixel in described direct picture; According to the pixel value of each second point and the magnitude relationship of the second predetermined prospect threshold value, determine the part foreground pixel in described verso images; According to the magnitude relationship of each pixel value of first and the first predetermined background threshold, determine the part background pixel in described direct picture; According to the pixel value of each second point and the magnitude relationship of the second predetermined background threshold, determine the part background pixel in described verso images.
In one embodiment, partial pixel determining device 32 comprises positive and negative comparing unit, this positive and negative comparing unit is configured to: according to the magnitude relationship of each pixel value of first with the pixel value of corresponding second point, determine the part foreground pixel in described direct picture and verso images and part background pixel.
In one embodiment, positive and negative comparing unit is further configured to: the business of the pixel value of described first and the pixel value of corresponding second point and the 3rd predetermined threshold value and the 4th threshold value are made comparisons, described 3rd threshold value is less than 1, and described 4th threshold value is greater than 1; If described business is less than described 3rd threshold value, then judge that be foreground pixel in described direct picture at described first, described second point is the background pixel in described verso images; If described business is greater than described 4th threshold value, then judge that described second point is the foreground pixel in described verso images, described first is background pixel in described direct picture.
In one embodiment, type determination device 34 is further configured to: the average pixel value calculating described four models respectively to and variance; According to the average pixel value of described four models to variance, pixel value that this pixel is right, calculate this pixel respectively to the distance to described four models; Type right for this pixel is defined as the type that the pixel of model corresponding to minor increment is right.
In one embodiment, the described direct picture that described dual-side image centering comprises and described verso images are gray level images, and described pixel value is gray-scale value.
In one embodiment, the described direct picture that described dual-side image centering comprises and described verso images are coloured images, and described pixel value is the value of each passage.
In one embodiment, decision maker 35 is further configured to: is judged to be the prospect on described direct picture at first that is confirmed as foreground pixel, the second point being confirmed as foreground pixel is judged to be the prospect on described verso images.
Below, with reference to Fig. 4 description saturating checkout equipment of the back of the body according to a second embodiment of the present invention.
Fig. 4 shows the block diagram of the saturating checkout equipment of the back of the body according to a second embodiment of the present invention.As shown in Figure 4, the saturating checkout equipment 400 of this back of the body of the saturating checkout equipment of the back of the body according to a second embodiment of the present invention 400 comprises: acquisition device 31, be configured to: the direct picture and the verso images that obtain the dual-side image centering of having alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel pair formed with corresponding described second point; Partial pixel determining device 32, is configured to: determine the part foreground pixel in described direct picture and described verso images and part background pixel; Candidate's prospect determining device 40, is configured to determine the candidate's foreground pixel in described direct picture and described verso images; Model building device 33, be configured to: based on the determination result of described partial pixel determining device 32, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel; Type determination device 44, be configured to: in described candidate's foreground pixel, pixel pair belonging to pixel except described part foreground pixel, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models, respectively to determine the type that this pixel is right; Decision maker 35, be configured to: the back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture is saturating.
In one embodiment, candidate's prospect determining device 40 is further configured to: if the pixel value of described first is less than or equal to the first predetermined background threshold, then determine that be candidate's foreground pixel in described direct picture at described first; If the pixel value of described second point is less than or equal to the second predetermined background threshold, then determine that described second point is the candidate's foreground pixel in described verso images.
Because the process in each device included in the saturating checkout equipment of the back of the body according to the present invention and unit is similar with the process in included each step in the saturating detection method of the above-described back of the body respectively, therefore for simplicity, the detailed description of these devices and unit is omitted at this.
In addition, still need here it is noted that each component devices, unit can be configured by software, firmware, hardware or its mode combined in the said equipment.Configure spendable concrete means or mode is well known to those skilled in the art, do not repeat them here.When being realized by software or firmware, to the computing machine (multi-purpose computer 500 such as shown in Fig. 5) with specialized hardware structure, the program forming this software is installed from storage medium or network, this computing machine, when being provided with various program, can perform various functions etc.
Fig. 5 shows and can be used for implementing the schematic block diagram according to the computing machine of the method and apparatus of the embodiment of the present invention.
In Figure 5, CPU (central processing unit) (CPU) 501 performs various process according to the program stored in ROM (read-only memory) (ROM) 502 or from the program that storage area 508 is loaded into random access memory (RAM) 503.In RAM503, also store the data required when CPU501 performs various process etc. as required.CPU501, ROM502 and RAM503 are connected to each other via bus 504.Input/output interface 505 is also connected to bus 504.
Following parts are connected to input/output interface 505: importation 506(comprises keyboard, mouse etc.), output 507(comprises display, such as cathode-ray tube (CRT) (CRT), liquid crystal display (LCD) etc., and loudspeaker etc.), storage area 508(comprises hard disk etc.), communications portion 509(comprises network interface unit such as LAN card, modulator-demodular unit etc.).Communications portion 509 is via network such as the Internet executive communication process.As required, driver 510 also can be connected to input/output interface 505.Detachable media 511 such as disk, CD, magneto-optic disk, semiconductor memory etc. can be installed on driver 510 as required, and the computer program therefrom read is installed in storage area 508 as required.
When series of processes above-mentioned by software simulating, from network such as the Internet or storage medium, such as detachable media 511 installs the program forming software.
It will be understood by those of skill in the art that this storage medium is not limited to wherein having program stored therein shown in Fig. 5, distributes the detachable media 511 to provide program to user separately with equipment.The example of detachable media 511 comprises disk (comprising floppy disk (registered trademark)), CD (comprising compact disc read-only memory (CD-ROM) and digital universal disc (DVD)), magneto-optic disk (comprising mini-disk (MD) (registered trademark)) and semiconductor memory.Or hard disk that storage medium can be ROM502, comprise in storage area 508 etc., wherein computer program stored, and user is distributed to together with comprising their equipment.
The present invention also proposes a kind of program product storing the instruction code of machine-readable.When described instruction code is read by machine and performs, the above-mentioned method according to the embodiment of the present invention can be performed.
Correspondingly, be also included within of the present invention disclosing for carrying the above-mentioned storage medium storing the program product of the instruction code of machine-readable.Described storage medium includes but not limited to floppy disk, CD, magneto-optic disk, storage card, memory stick etc.
Above in the description of the specific embodiment of the invention, the feature described for a kind of embodiment and/or illustrate can use in one or more other embodiment in same or similar mode, combined with the feature in other embodiment, or substitute the feature in other embodiment.
Should emphasize, term " comprises/comprises " existence referring to feature, key element, step or assembly when using herein, but does not get rid of the existence or additional of one or more further feature, key element, step or assembly.
In addition, method of the present invention be not limited to specifications in describe time sequencing perform, also can according to other time sequencing ground, perform concurrently or independently.Therefore, the execution sequence of the method described in this instructions is not construed as limiting technical scope of the present invention.
Although above by the description of specific embodiments of the invention to invention has been disclosure, should be appreciated that, above-mentioned all embodiments and example are all illustrative, and not restrictive.Those skilled in the art can design various amendment of the present invention, improvement or equivalent in the spirit and scope of claims.These amendments, improvement or equivalent also should be believed to comprise in protection scope of the present invention.
Remarks
1. carry on the back a saturating detection method, the saturating detection method of this back of the body comprises:
Obtain direct picture and the verso images of the dual-side image centering of having alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel pair formed with corresponding described second point;
Determine the part foreground pixel in described direct picture and described verso images and part background pixel;
Based on above-mentioned determination result, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel;
For the pixel pair be not modeled, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models, respectively to determine the type that this pixel is right;
The back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture is saturating.
2. the method as described in remarks 1, wherein saidly determine that part foreground pixel in direct picture and verso images and part background pixel comprise:
According to the magnitude relationship of each pixel value of first and the first predetermined prospect threshold value, determine the part foreground pixel in described direct picture;
According to the pixel value of each second point and the magnitude relationship of the second predetermined prospect threshold value, determine the part foreground pixel in described verso images;
According to the magnitude relationship of each pixel value of first and the first predetermined background threshold, determine the part background pixel in described direct picture;
According to the pixel value of each second point and the magnitude relationship of the second predetermined background threshold, determine the part background pixel in described verso images.
3. the method as described in remarks 1 or 2, wherein saidly determine that the part foreground pixel in direct picture and verso images and part background pixel comprise: according to the magnitude relationship of each pixel value of first with the pixel value of corresponding second point, determine the part foreground pixel in described direct picture and verso images and part background pixel.
4. the method as described in remarks 3, wherein said according to the magnitude relationship of each pixel value of first with the pixel value of corresponding second point, determine that part foreground pixel in direct picture and verso images and part background pixel comprise:
The business of the pixel value of described first and the pixel value of corresponding second point and the 3rd predetermined threshold value and the 4th threshold value are made comparisons, described 3rd threshold value is less than 1, and described 4th threshold value is greater than 1;
If described business is less than described 3rd threshold value, then described first is foreground pixel in described direct picture, and described second point is the background pixel in described verso images;
If described business is greater than described 4th threshold value, then described second point is the foreground pixel in described verso images, and described first is background pixel in described direct picture.
5. the method as described in remarks 1, also comprises: determine the candidate's foreground pixel in described direct picture and described verso images;
The wherein said pixel be not modeled is to comprising: in described candidate's foreground pixel, except described part foreground pixel the pixel pair belonging to pixel.
6. the method as described in remarks 5, the wherein said candidate's foreground pixel determined in direct picture and verso images comprises:
If the pixel value of described first is less than or equal to the first predetermined background threshold, then determine that be candidate's foreground pixel in described direct picture at described first;
If the pixel value of described second point is less than or equal to the second predetermined background threshold, then determine that described second point is the candidate's foreground pixel in described verso images.
7. the method as described in remarks 1, the wherein said pixel pair for not being modeled, the predetermined characteristic right according to this pixel, calculate the similarity of this pixel pair and four models respectively, to determine this pixel, right type comprises:
The average pixel value calculating described four models respectively to and variance;
According to the average pixel value of described four models to variance, pixel value that this pixel is right, calculate this pixel respectively to the distance to described four models;
Type right for this pixel is defined as the type that the pixel of model corresponding to minor increment is right.
8. the method as described in remarks 1, the described direct picture that wherein said dual-side image centering comprises and described verso images are gray level images, and described pixel value is gray-scale value.
9. the method as described in remarks 1, the described direct picture that wherein said dual-side image centering comprises and described verso images are coloured images, and described pixel value is the value of each passage.
10. the method as described in remarks 1, also comprises:
Is judged to be the prospect on described direct picture at first that is confirmed as foreground pixel, the second point being confirmed as foreground pixel is judged to be the prospect on described verso images.
11. 1 kinds of saturating checkout equipments of the back of the body, the saturating checkout equipment of this back of the body comprises:
Acquisition device, be configured to: the direct picture and the verso images that obtain the dual-side image centering of having alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel pair formed with corresponding described second point;
Partial pixel determining device, is configured to: determine the part foreground pixel in described direct picture and described verso images and part background pixel;
Model building device, be configured to: based on the determination result of described partial pixel determining device, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel;
Type determination device, is configured to: for the pixel pair be not modeled, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models respectively, to determine the type that this pixel is right;
Decision maker, be configured to: the back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture is saturating.
12. saturating checkout equipments of the back of the body as described in remarks 11, wherein said partial pixel determining device comprises threshold decision unit, and this threshold decision unit is configured to:
According to the magnitude relationship of each pixel value of first and the first predetermined prospect threshold value, determine the part foreground pixel in described direct picture;
According to the pixel value of each second point and the magnitude relationship of the second predetermined prospect threshold value, determine the part foreground pixel in described verso images;
According to the magnitude relationship of each pixel value of first and the first predetermined background threshold, determine the part background pixel in described direct picture;
According to the pixel value of each second point and the magnitude relationship of the second predetermined background threshold, determine the part background pixel in described verso images.
13. saturating checkout equipments of the back of the body as described in remarks 11 or 12, wherein said partial pixel determining device comprises positive and negative comparing unit, this positive and negative comparing unit is configured to: according to the magnitude relationship of each pixel value of first with the pixel value of corresponding second point, determine the part foreground pixel in described direct picture and verso images and part background pixel.
14. saturating checkout equipments of the back of the body as described in remarks 13, wherein said positive and negative comparing unit is further configured to:
The business of the pixel value of described first and the pixel value of corresponding second point and the 3rd predetermined threshold value and the 4th threshold value are made comparisons, described 3rd threshold value is less than 1, and described 4th threshold value is greater than 1;
If described business is less than described 3rd threshold value, then judge that be foreground pixel in described direct picture at described first, described second point is the background pixel in described verso images;
If described business is greater than described 4th threshold value, then judge that described second point is the foreground pixel in described verso images, described first is background pixel in described direct picture.
15. saturating checkout equipments of the back of the body as described in remarks 11, also comprise: candidate's prospect determining device, are configured to determine the candidate's foreground pixel in described direct picture and described verso images;
The wherein said pixel be not modeled is to comprising: in described candidate's foreground pixel, except described part foreground pixel the pixel pair belonging to pixel.
16. saturating checkout equipments of the back of the body as described in remarks 15, wherein said candidate's prospect determining device is further configured to:
If the pixel value of described first is less than or equal to the first predetermined background threshold, then determine that be candidate's foreground pixel in described direct picture at described first;
If the pixel value of described second point is less than or equal to the second predetermined background threshold, then determine that described second point is the candidate's foreground pixel in described verso images.
17. saturating checkout equipments of the back of the body as described in remarks 11, wherein said type determination device is further configured to:
The average pixel value calculating described four models respectively to and variance;
According to the average pixel value of described four models to variance, pixel value that this pixel is right, calculate this pixel respectively to the distance to described four models;
Type right for this pixel is defined as the type that the pixel of model corresponding to minor increment is right.
18. saturating checkout equipments of the back of the body as described in remarks 11, the described direct picture that wherein said dual-side image centering comprises and described verso images are gray level images, and described pixel value is gray-scale value.
19. saturating checkout equipments of the back of the body as described in remarks 11, the described direct picture that wherein said dual-side image centering comprises and described verso images are coloured images, and described pixel value is the value of each passage.
20. saturating checkout equipments of the back of the body as described in remarks 11, wherein said decision maker is further configured to:
Is judged to be the prospect on described direct picture at first that is confirmed as foreground pixel, the second point being confirmed as foreground pixel is judged to be the prospect on described verso images.
Claims (10)
1. carry on the back a saturating detection method, the saturating detection method of this back of the body comprises:
Obtain direct picture and the verso images of the dual-side image centering of having alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel pair formed with corresponding described second point;
Determine the part foreground pixel in described direct picture and described verso images and part background pixel;
Based on above-mentioned determination result, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel;
For the pixel pair be not modeled, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models, respectively to determine the type that this pixel is right;
The back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture is saturating.
2. the method for claim 1, wherein saidly determine that part foreground pixel in direct picture and verso images and part background pixel comprise:
According to the magnitude relationship of each pixel value of first and the first predetermined prospect threshold value, determine the part foreground pixel in described direct picture;
According to the pixel value of each second point and the magnitude relationship of the second predetermined prospect threshold value, determine the part foreground pixel in described verso images;
According to the magnitude relationship of each pixel value of first and the first predetermined background threshold, determine the part background pixel in described direct picture;
According to the pixel value of each second point and the magnitude relationship of the second predetermined background threshold, determine the part background pixel in described verso images.
3. method as claimed in claim 1 or 2, wherein saidly determine that the part foreground pixel in direct picture and verso images and part background pixel comprise: according to the magnitude relationship of each pixel value of first with the pixel value of corresponding second point, determine the part foreground pixel in described direct picture and verso images and part background pixel.
4. method as claimed in claim 3, wherein said according to the magnitude relationship of each pixel value of first with the pixel value of corresponding second point, determines that part foreground pixel in direct picture and verso images and part background pixel comprise:
The business of the pixel value of described first and the pixel value of corresponding second point and the 3rd predetermined threshold value and the 4th threshold value are made comparisons, described 3rd threshold value is less than 1, and described 4th threshold value is greater than 1;
If described business is less than described 3rd threshold value, then described first is foreground pixel in described direct picture, and described second point is the background pixel in described verso images;
If described business is greater than described 4th threshold value, then described second point is the foreground pixel in described verso images, and described first is background pixel in described direct picture.
5. the method for claim 1, also comprises: determine the candidate's foreground pixel in described direct picture and described verso images;
The wherein said pixel be not modeled is to comprising: in described candidate's foreground pixel, except described part foreground pixel the pixel pair belonging to pixel.
6. method as claimed in claim 5, the wherein said candidate's foreground pixel determined in direct picture and verso images comprises:
If the pixel value of described first is less than or equal to the first predetermined background threshold, then determine that be candidate's foreground pixel in described direct picture at described first;
If the pixel value of described second point is less than or equal to the second predetermined background threshold, then determine that described second point is the candidate's foreground pixel in described verso images.
7. the method for claim 1, the wherein said pixel pair for not being modeled, the predetermined characteristic right according to this pixel, calculate the similarity of this pixel pair and four models respectively, to determine this pixel, right type comprises:
The average pixel value calculating described four models respectively to and variance;
According to the average pixel value of described four models to variance, pixel value that this pixel is right, calculate this pixel respectively to the distance to described four models;
Type right for this pixel is defined as the type that the pixel of model corresponding to minor increment is right.
8. the method for claim 1, also comprises:
Is judged to be the prospect on described direct picture at first that is confirmed as foreground pixel, the second point being confirmed as foreground pixel is judged to be the prospect on described verso images.
9. carry on the back a saturating checkout equipment, the saturating checkout equipment of this back of the body comprises:
Acquisition device, be configured to: the direct picture and the verso images that obtain the dual-side image centering of having alignd, wherein for the known second point corresponding with it in described verso images of each in described direct picture first, thus obtain by described first pixel pair formed with corresponding described second point;
Partial pixel determining device, is configured to: determine the part foreground pixel in described direct picture and described verso images and part background pixel;
Model building device, be configured to: based on the determination result of described partial pixel determining device, pixel for following Four types carries out modeling to based on predetermined characteristic, to form four models: first and second point are all foreground pixels, first and second point are all background pixel, first be foreground pixel and second point be background pixel, first be background pixel and second point is foreground pixel;
Type determination device, is configured to: for the pixel pair be not modeled, the predetermined characteristic right according to this pixel, calculates the similarity of this pixel pair and described four models respectively, to determine the type that this pixel is right;
Decision maker, be configured to: the back of the body be judged to be on described verso images by second point that is corresponding with first that is confirmed as foreground pixel, that be confirmed as background pixel is saturating, by corresponding with the second point being confirmed as foreground pixel, be confirmed as background pixel first be judged to be that the back of the body on described direct picture is saturating.
10. the saturating checkout equipment of the back of the body as claimed in claim 9, also comprises: candidate's prospect determining device, is configured to determine the candidate's foreground pixel in described direct picture and described verso images;
The wherein said pixel be not modeled is to comprising: in described candidate's foreground pixel, except described part foreground pixel the pixel pair belonging to pixel.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310744057.8A CN104751165A (en) | 2013-12-30 | 2013-12-30 | Back-through detection method and device |
| JP2014262518A JP2015130166A (en) | 2013-12-30 | 2014-12-25 | Rear surface watermark detection method and rear surface water mark detection device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310744057.8A CN104751165A (en) | 2013-12-30 | 2013-12-30 | Back-through detection method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104751165A true CN104751165A (en) | 2015-07-01 |
Family
ID=53590820
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310744057.8A Pending CN104751165A (en) | 2013-12-30 | 2013-12-30 | Back-through detection method and device |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP2015130166A (en) |
| CN (1) | CN104751165A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105631837A (en) * | 2015-10-27 | 2016-06-01 | 江苏思曼特信用管理有限公司 | Text image back penetration removing method |
| CN106780541A (en) * | 2016-12-28 | 2017-05-31 | 南京师范大学 | A kind of improved background subtraction method |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7210187B2 (en) | 2018-08-10 | 2023-01-23 | 株式会社ミマキエンジニアリング | LIQUID EJECTING APPARATUS AND LIQUID EJECTING METHOD |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050041883A1 (en) * | 2000-09-29 | 2005-02-24 | Maurer Ron P. | Method for enhancing compressibility and visual quality of scanned document images |
| CN1700253A (en) * | 2004-05-20 | 2005-11-23 | 明基电通股份有限公司 | Image correction system and method, and computer readable storage medium |
-
2013
- 2013-12-30 CN CN201310744057.8A patent/CN104751165A/en active Pending
-
2014
- 2014-12-25 JP JP2014262518A patent/JP2015130166A/en not_active Withdrawn
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050041883A1 (en) * | 2000-09-29 | 2005-02-24 | Maurer Ron P. | Method for enhancing compressibility and visual quality of scanned document images |
| CN1700253A (en) * | 2004-05-20 | 2005-11-23 | 明基电通股份有限公司 | Image correction system and method, and computer readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| ERIC DUBOIS等: "Joint Compression and Restoration of Documents with Bleed-through", 《IS&T ARCHIVING》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105631837A (en) * | 2015-10-27 | 2016-06-01 | 江苏思曼特信用管理有限公司 | Text image back penetration removing method |
| CN106780541A (en) * | 2016-12-28 | 2017-05-31 | 南京师范大学 | A kind of improved background subtraction method |
| CN106780541B (en) * | 2016-12-28 | 2019-06-14 | 南京师范大学 | A kind of improved background subtraction method |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2015130166A (en) | 2015-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bazin et al. | Globally optimal line clustering and vanishing point estimation in manhattan world | |
| CN112581463A (en) | Image defect detection method and device, electronic equipment, storage medium and product | |
| Huh et al. | Detecting conjugate-point pairs for map alignment between two polygon datasets | |
| CN105488515A (en) | Method for training convolutional neural network classifier and image processing device | |
| CN112966742A (en) | Model training method, target detection method and device and electronic equipment | |
| CN114677565B (en) | Training method and image processing method and device for feature extraction network | |
| Courtney et al. | Algorithmic modelling for performance evaluation | |
| CN105069454A (en) | Image identification method and apparatus | |
| CN111461164A (en) | Sample data set capacity expansion method and model training method | |
| CN111783844A (en) | Target detection model training method and device based on deep learning and storage medium | |
| Fazan et al. | Rectilinear building roof contour extraction based on snakes and dynamic programming | |
| CN104751165A (en) | Back-through detection method and device | |
| CN113032862A (en) | Building information model checking method and device and terminal equipment | |
| CN114972947B (en) | Depth scene text detection method and device based on fuzzy semantic modeling | |
| CN104077765A (en) | Image segmentation device, image segmentation method and program | |
| CN102855635A (en) | Method and device for determining human body action cycles and recognizing human body actions | |
| CN101833760A (en) | Background modeling method and device based on image blocks | |
| CN104700388A (en) | Method and device for extracting distorted lines from images | |
| CN114913321A (en) | Object attention mining method and system based on local-to-global knowledge migration | |
| CN102855462A (en) | Method and device for determining human body action cycle and recognizing human body action | |
| CN102855477B (en) | Method and device for recognizing direction of characters in image block | |
| CN103837135A (en) | Workpiece detecting method and system | |
| CN103426140A (en) | Method and apparatus for image tamper-proofing verification, and image tamper-proofing method and system | |
| CN103136536A (en) | System and method for detecting target and method for exacting image features | |
| CN104200222A (en) | Picture object identifying method based on factor graph model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150701 |
|
| RJ01 | Rejection of invention patent application after publication |