CN103942543A - Image recognition method and device - Google Patents
Image recognition method and device Download PDFInfo
- Publication number
- CN103942543A CN103942543A CN201410179080.1A CN201410179080A CN103942543A CN 103942543 A CN103942543 A CN 103942543A CN 201410179080 A CN201410179080 A CN 201410179080A CN 103942543 A CN103942543 A CN 103942543A
- Authority
- CN
- China
- Prior art keywords
- subimage
- image
- coding
- center
- pixel value
- 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
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention belongs to the technical field of image processing, and provides an image recognition method and device. The method comprises the steps that the gray value image of an image to be recognized is acquired; the gray value image is divided into a plurality of sub-images with the equal size according to a row and column equating mode; one or more sub-images are selected from the sub-images, the pixel values of one or more sub-images adjacent to the selected one or more sub-images are compared with the selected one or more sub-images, a code is determined according to the comparison result, and the code is matched with a pre-stored image code to recognize the image to be recognized. The code acquired based on pixel value comparison is adopted as the characteristic information of the image to be recognized to recognize the image to be recognized, and the accuracy rate of image recognition can be effectively improved.
Description
Technical field
The invention belongs to technical field of image processing, relate in particular to a kind of image-recognizing method and device.
Background technology
Existing image recognition is mostly on the basis that the yardstick based on consecutive image and image itself is constant, image to be identified.But in actual application, particularly, on TV, the TV of different model or different brands can do certain proportion, yardstick convergent-divergent to the image of inputting in Set Top Box, and image quality is changed.And user's hobby difference, the aspect ratio (for example 16:9 or 4:3) of selecting in watching TV also may be different, and the change of aspect ratio can cause content, the structure generation deformation of image itself equally.Adopt present image-recognizing method (such as SIFT feature point detection, Harris method or the method based on color image piecemeal etc.), in the time that graphical rule changes, identification that cannot be accurate and effective.
Summary of the invention
The object of the embodiment of the present invention is to provide a kind of image-recognizing method and device, to solve prior art in the time that graphical rule changes, and problem that cannot accurate and effective identification.
The embodiment of the present invention is achieved in that a kind of image-recognizing method, and described method comprises:
Obtain the gray-value image of image to be identified;
Described gray-value image is become to multiple equal-sized subimages by the mode division of row, column decile;
Selected one or more subimages from described subimage, for selected described one or more subimages, relatively itself and the pixel value of contiguous one or more subimages, and determine coding according to comparative result, described coding and pre-stored Image Coding are mated to identify described image to be identified.
Another object of the embodiment of the present invention is to provide a kind of pattern recognition device, and described device comprises:
Image acquisition unit, for obtaining the gray-value image of image to be identified;
Image division unit, for becoming multiple equal-sized subimages by described gray-value image by the mode division of row, column decile;
Coding unit, for selecting one or more subimages from described subimage, for selected described one or more subimages, relatively itself and the pixel value of contiguous one or more subimages, and determine coding according to comparative result, described coding and pre-stored Image Coding are mated to identify described image to be identified.
The beneficial effect that the embodiment of the present invention compared with prior art exists is: the embodiment of the present invention is by being divided into the gray-value image of individual image to be identified multiple equal-sized subimages, and the pixel value of every number of sub images one or more subimages adjacent thereto of relatively selecting, determine coding according to comparative result, the characteristic information using described coding as described characteristic to be identified mates to identify described image to be identified with pre-stored image.Due in the time that graphical rule changes, less on the impact that the impact of general image structural information distribution is less, therefore the coding relatively obtaining based on described pixel value can be used as the effective characteristic information of described image to be identified, thereby ensures the accuracy rate of individual image recognition.
Brief description of the drawings
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the realization flow figure of the image-recognizing method that provides of first embodiment of the invention;
Fig. 2 a in Fig. 2 is that exemplary plot, Fig. 2 b of the 16:9 aspect ratio that provides of first embodiment of the invention is the exemplary plot of the 4:3 aspect ratio that provides of first embodiment of the invention;
Fig. 3 is the exemplary plot of the image interception that provides of first embodiment of the invention;
Fig. 4 is the exemplary plot of the Image Coding that provides of first embodiment of the invention;
Fig. 5 is the composition structural drawing of the pattern recognition device that provides of second embodiment of the invention.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
For technical solutions according to the invention are described, describe below by specific embodiment.
embodiment mono-:
Fig. 1 shows the realization flow of the image-recognizing method that the first embodiment provides, and details are as follows for the method process:
In step S101, obtain the gray-value image of image to be identified.
In the present embodiment, in order to reduce original image data amount, improve efficiency and the accuracy rate of image recognition, terminal device (for example intelligent TV set) is getting after image to be identified, first judge whether described image to be identified is gray-value image, if not, described image to be identified is converted to gray-value image.
In step S102, described gray-value image is become to multiple equal-sized subimages by the mode division of row, column decile.
Concrete can be, from described gray-value image, intercept the image of two pre-sizings, the image of described two pre-sizings is pressed respectively to the pattern of row 2 × N+1 decile, row 2 × N+1 decile, be divided into (2 × N+1) × (2 × N+1) number of sub images, and the subimage of image to pre-sizing every described is numbered by its place ranks, described N is more than or equal to 1 integer.
It should be noted that, due to user's hobby difference, the aspect ratio (for example 16:9 or 4:3) of selecting in watching TV also may be different, and the change of aspect ratio can cause content, the structure generation deformation of image itself.Therefore,, in order to ensure the accuracy rate of image recognition, the present embodiment intercepts the image of two pre-sizings, the height H of the image of described two pre-sizings centered by the central point of described gray-value image
-getit is identical,
The relational expression of width is W
4-3-get=(W
16-9-get× W
16-9)/W
4-3
Wherein, W
16-9-getfor width, the W of first image intercepting
4-3-getfor the width of second image intercepting, W
16-9, W
4-3for fixed value, as shown in Fig. 2 a, 2b, in television image, the height of the height of the aspect ratio of 16:9 and the aspect ratio of 4:3 is identical, i.e. H
16-9=H
4-3=H, the width W of the aspect ratio of 16:9
16-9width W with the aspect ratio of 4:3
4-3for known fixed value.
It should be noted that, the present embodiment using described first image the image under 16:9 aspect ratio as example.Because first picture altitude, the width that intercept are known, the height of second image (being the image under 4:3 aspect ratio) is identical with the height of first image, can obtain the width (as shown in Figure 3) of second image according to the relational expression of above-mentioned width.Equally, if the image under 4:3 aspect ratio, as example, can obtain the image under 4:3 aspect ratio equally according to the relational expression of above-mentioned width using described first image.
The present embodiment adopts aforesaid way in an image to be identified, to intercept the image of two different pictures ratios, thereby ensure to be switched to after different aspect ratios user, still can from truncated picture, obtain the image consistent with pre-stored image ratio, ensure the accuracy rate of image recognition.
In step S103, selected one or more subimages from described subimage, for selected described one or more subimages, relatively itself and the pixel value of contiguous one or more subimages, and determine coding according to comparative result, described coding and pre-stored Image Coding are mated to identify described image to be identified.
In the present embodiment, described from described subimage selected one or more subimages, for selected described one or more subimages, relatively itself and the pixel value of contiguous one or more subimages, and determine that according to comparative result coding comprises:
Step 1: from described subimage selected ((N+1), (N+1)) subimage, and by described ((N+1), (N+1)) subimage is as center subimage, wherein said ((N+1), (N+1)) is the row and column at subimage place;
Step 2: Jiang Gai center subimage subimage adjacent thereto carries out pixel value comparison, the pixel value of Ruo Gai center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
Step 3: subimage contiguous every number of sub images in Jiang Gai center is as center subimage, and return to step 2 and carry out, until the every number of sub images in the image of described pre-sizing is all completeer;
Step 4: the coding that repeats relatively to obtain in delete step 2 and step 3, and arrange remaining coding by predefined procedure, coding after arranging is as the coding of described image to be identified, is wherein saidly remainingly encoded to the coding that step 2 and step 3 obtain and deducts the coding that repeats relatively to obtain.
Embodiment for a better understanding of the present invention, illustrates above-mentioned cataloged procedure below by a concrete example:
As shown in Figure 4, step 1: truncated picture is divided into 5 × 5 (being N=2) number of sub images, and every number of sub images is numbered by its place ranks, institute's call number is from (1,1) to (5,5) as shown in Figure 4;
Step 2: with (3, 3) centered by the subimage of numbering, respectively eight number of sub images adjacent thereto ((2, 2), (2, 3), (2, 4), (3, 2), (3, 4), (4, 2), (4, 3), (4, 4) subimage of numbering) carry out pixel value comparison, if the pixel value of center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0 (for example (3, 3) numbering subimage pixel value and be greater than (2, 2) numbering subimage pixel value and, be numbered 1), in step 2, obtain 8 codings,
Step 3: with (2,2), (2,3), (2,4), (3,2), (3,4), (4,2), (4,3), (4,4) centered by the subimage of numbering, and subimage adjacent thereto compares respectively, if the pixel value of center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0; For example (2,2) its contiguous subimage of the subimage of numbering comprises (1,1), (1,2), (1,3), (2,1), (2,3), (3,1), the subimage of (3,2), (3,3) numbering; Its contiguous subimage of the subimage of (2,3) numbering comprises (1,2), (1,3), (1,4), (2,2), (2,4), (3,2), (3,3), (3,4); In step 3, obtain 64 codings;
Step 4: the coding (coding that repeats comparison has 20) that repeats relatively to obtain in delete step 2 and step 3, and arrange remaining coding (8+64-20=52 position) by sequencing relatively, coding after arranging is as the coding of described image to be identified, is wherein saidly remainingly encoded to the coding that step 2 and step 3 obtain and deducts the coding that repeats relatively to obtain.
As another preferable examples of the present invention, described from described subimage selected one or more subimages, for selected described one or more subimages, relatively itself and the pixel value of contiguous one or more subimages, and determine that according to comparative result coding comprises:
Step 1: from described subimage selected ((N+1), (N+1)) subimage, and by described ((N+1), (N+1)) subimage is as the first center subimage, wherein said ((N+1), (N+1)) is the row and column at subimage place;
Step 2: this first center subimage subimage adjacent thereto is carried out to pixel value comparison, if the pixel value of this first center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
Step 3: every number of sub images that this first center subimage is close to angular direction is as the second center subimage, and described the second center subimage subimage adjacent thereto is compared, if the pixel value of described the second center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
Step 4: using every number of sub images of this first center subimage vertical direction and horizontal direction vicinity as San center subimage, and the perpendicular direction of described San center subimage and the contiguous subimage of horizontal direction are compared, if the pixel value of center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
Step 5: using every number of sub images of every number of sub images contiguous described the second center subimage and San center subimage vertical direction and horizontal direction vicinity as the first center subimage, and return to execution step 2, until the every number of sub images in the image of described pre-sizing is all completeer;
Step 6: delete step 2 is to the coding that repeats relatively to obtain in step 5, and arrange remaining coding by predefined procedure, coding after arranging is as the coding of described image to be identified, is wherein saidly remainingly encoded to the coding that step 2 to step 5 obtains and deducts the coding that repeats relatively to obtain.
Embodiment for a better understanding of the present invention, illustrates above-mentioned cataloged procedure below by a concrete example:
As shown in Figure 4, step 1: truncated picture is divided into 5 × 5 (being N=2) number of sub images, and every number of sub images is numbered by its place ranks, institute's call number is from (1,1) to (5,5) as shown in Figure 4;
Step 2: with (3, 3) centered by the subimage of numbering, respectively eight number of sub images adjacent thereto ((2, 2), (2, 3), (2, 4), (3, 2), (3, 4), (4, 2), (4, 3), (4, 4) subimage of numbering) carry out pixel value comparison, if the pixel value of center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0 (for example (3, 3) numbering subimage pixel value and be greater than (2, 2) numbering subimage pixel value and, be numbered 1), in step 2, obtain 8 codings,
Step 3: with (2,2), (2,4), (4,2), (4,4) centered by the subimage of numbering, and subimage adjacent thereto compares respectively, if the pixel value of center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0; For example (2,2) its contiguous subimage of the subimage of numbering comprises (1,1), (1,2), (1,3), (2,1), (2,3), (3,1), the subimage of (3,2), (3,3) numbering; In step 3, obtain 32 codings;
Step 4: with the (2,3), (3,2), (3,4), (4,3), centered by subimage, the contiguous subimage of perpendicular direction and horizontal direction compares respectively, if the pixel value of center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0; The contiguous subimage of its horizontal direction of subimage of for example (2,3) numbering is (2,2), (2,4), and the contiguous subimage of vertical direction is (1,3), (3,3); In step 4, obtain 16 codings;
Step 5: the coding that repeats relatively to obtain in delete step 2, step 3 and step 4 (for example comparison of (2,2), (3,3) in the comparison of (3,3), (2,2) and step 2 in step 1; In step 1 (3,3), (2,3) and in step 2 (2,2), (2,3) with step 3 in (2,3), (3,3), the comparison of (2,3), (2,2), totally 16 of the codings that repeats relatively to obtain), and arrange remaining coding (8+32+16-16=40 position) by sequencing relatively, the coding after arranging is as the coding of described image to be identified, is wherein saidly remainingly encoded to the coding that step 2, step 3 and step 4 obtain and deducts the coding that repeats relatively to obtain.
It should be noted that, two images that the present embodiment intercepts all need to adopt above-mentioned coded system to obtain corresponding coding, and the corresponding coding obtaining and pre-stored Image Coding are mated to identify described image to be identified.If pre-stored is the Image Coding of 4:3, the coding corresponding with the image of 4:3 intercepting mates, and be the Image Coding of 16:9 if pre-stored, and the coding corresponding with the image of the 16:9 of intercepting mates.
The characteristic information of the embodiment of the present invention using the coding relatively obtaining based on described pixel value as described image to be identified identified described image to be identified, can effectively improve the accuracy rate of image recognition.And, by intercept the image of two different pictures ratios in an image to be identified, thereby ensure to be switched to after different aspect ratio (as 16:9 or 4:3) user, still can from truncated picture, obtain the image consistent with pre-stored image ratio, ensure higher image recognition rate.Described in the present embodiment, scheme especially has good recognition effect for image dwindling or amplify in 8% scope, compresses in the horizontal and vertical directions and move also to have stronger robustness for image simultaneously.
embodiment bis-:
Fig. 5 shows the composition structure of the pattern recognition device that second embodiment of the invention provides, and for convenience of explanation, only shows the part relevant to the embodiment of the present invention.
This pattern recognition device can be to run on the unit that such as, software unit, hardware cell or software and hardware in each terminal device (mobile phone, panel computer, intelligent TV set etc.) combine, and also can be used as independently suspension member and is integrated in described terminal device or runs in the application system of described terminal device.
This pattern recognition device comprises image acquisition unit 51, image division unit 52 and coding unit 53.Wherein, the concrete function of each unit is as follows:
Image acquisition unit 51, for obtaining the gray-value image of image to be identified;
Image division unit 52, for becoming multiple equal-sized subimages by described gray-value image by the mode division of row, column decile;
Coding unit 53, for selecting one or more subimages from described subimage, for selected described one or more subimages, relatively itself and the pixel value of contiguous one or more subimages, and determine coding according to comparative result, described coding and pre-stored Image Coding are mated to identify described image to be identified.
Further, described image division unit 52 specifically for:
From described gray-value image, intercept the image of two pre-sizings, the image of described two pre-sizings is pressed respectively to the pattern of row 2 × N+1 decile, row 2 × N+1 decile, be divided into (2 × N+1) × (2 × N+1) number of sub images, and the subimage of image to pre-sizing every described is numbered by its place ranks, described N is more than or equal to 1 integer.
Further, described coding unit 53 comprises:
The first coding module 531, for selecting ((N+1) from described subimage, (N+1)) subimage, and by described ((N+1), (N+1)) subimage is as center subimage, wherein said ((N+1), (N+1)) is the row and column at subimage place;
The second coding module 532, carries out pixel value comparison for Jiang Gai center subimage subimage adjacent thereto, the pixel value of Ruo Gai center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
The 3rd coding module 533, every number of sub images of being close to for Jiang Gai center subimage is as center subimage;
The first control module 534, carries out for the circulation of controlling the second coding module and the 3rd coding module, until the every number of sub images in the image of described pre-sizing is all completeer;
The 4th coding module 535, for deleting the coding that repeats relatively to obtain, and arrange remaining coding by predefined procedure, coding after arranging is as the coding of described image to be identified, is wherein saidly remainingly encoded to the coding that the second coding module 532 obtains and deducts the coding that repeats relatively to obtain.
Further, described coding module 53 comprises:
The 5th coding module 536, for selecting ((N+1) from described subimage, (N+1)) subimage, and by described ((N+1), (N+1)) subimage is as the first center subimage, wherein said ((N+1), (N+1)) is the row and column at subimage place;
The 6th coding module 537, for this first center subimage subimage adjacent thereto is carried out to pixel value comparison, if the pixel value of this first center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
The 7th coding module 538, the every number of sub images that is used for this first center subimage to be close to angular direction is as the second center subimage, and described the second center subimage subimage adjacent thereto is compared, if the pixel value of described the second center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
The 8th coding module 539, be used for using every number of sub images of this first center subimage vertical direction and horizontal direction vicinity as San center subimage, and the perpendicular direction of described San center subimage and the contiguous subimage of horizontal direction are compared, if the pixel value of center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
The 9th coding module 5310, for using every number of sub images contiguous described the second center subimage and San center subimage vertical direction and the contiguous every number of sub images of horizontal direction as the first center subimage;
The second control module 5311, carries out for the circulation of controlling the 6th coding module, the 7th coding module, the 8th coding module and the 9th coding module, until the every number of sub images in the image of described pre-sizing is all completeer;
The tenth coding module 5312, for deleting the coding that repeats relatively to obtain, and arrange remaining coding by predefined procedure, coding after arranging is as the coding of described image to be identified, is wherein saidly remainingly encoded to the coding that the 6th coding module 537, the 7th coding module 538 and the 8th coding module 539 obtain and deducts the coding that repeats relatively to obtain.
Further, the height of the image of described two pre-sizings is identical,
The relational expression of width is W
4-3-get=(W
16-9-get× W
16-9)/W
4-3
Wherein, W
16-9, W
4-3for known value, W
16-9-getfor width, the W of first image intercepting
4-3-getfor the width of second image intercepting.
Those skilled in the art can be well understood to, for convenience and simplicity of description, only be illustrated with the division of above-mentioned each functional unit, module, in practical application, can above-mentioned functions be distributed and completed by different functional units or module as required, be divided into different functional units or module by the inner structure of device, to complete all or part of function described above.Each functional unit in embodiment, module can be integrated in a processing unit, also can be that the independent physics of unit exists, also can be integrated in a unit two or more unit, above-mentioned integrated unit both can adopt the form of hardware to realize, and also can adopt the form of SFU software functional unit to realize.In addition, the concrete title of each functional unit, module also, just for the ease of mutual differentiation, is not limited to the application's protection domain.The specific works process of unit, module in said apparatus, can, with reference to the corresponding process in preceding method embodiment, not repeat them here.
In sum, the embodiment of the present invention is by comparing the pixel value of subimage and its contiguous subimage, determine coding according to comparative result, characteristic information using the coding relatively obtaining based on described pixel value as described image to be identified is identified described image to be identified, can effectively improve the accuracy rate of image recognition.And, by intercept the image of two different pictures ratios in an image to be identified, thereby ensure to be switched to after different aspect ratio (as 16:9 or 4:3) user, still can from truncated picture, obtain the image consistent with pre-stored image ratio, ensure higher image recognition rate.Especially have good recognition effect for image dwindling or amplify in 8% scope, compress in the horizontal and vertical directions and move for image also has stronger robustness simultaneously, makes the embodiment of the present invention possess stronger ease for use and practicality.
Those of ordinary skill in the art it is also understood that, the all or part of step realizing in above-described embodiment method is can carry out the hardware that instruction is relevant by program to complete, described program can be being stored in a computer read/write memory medium, described storage medium, comprises ROM/RAM, disk, CD etc.
Above content is in conjunction with concrete preferred implementation further description made for the present invention, can not assert that specific embodiment of the invention is confined to these explanations.For general technical staff of the technical field of the invention; make without departing from the inventive concept of the premise some alternative or obvious modification that are equal to; and performance or purposes identical, all should be considered as belonging to the present invention by the definite scope of patent protection of submitted to claims.
Claims (10)
1. an image-recognizing method, is characterized in that, described method comprises:
Obtain the gray-value image of image to be identified;
Described gray-value image is become to multiple equal-sized subimages by the mode division of row, column decile;
Selected one or more subimages from described subimage, for selected described one or more subimages, relatively itself and the pixel value of contiguous one or more subimages, and determine coding according to comparative result, described coding and pre-stored Image Coding are mated to identify described image to be identified.
2. the method for claim 1, is characterized in that, describedly becomes multiple equal-sized subimages to comprise by the mode division of row, column decile described gray-value image:
From described gray-value image, intercept the image of two pre-sizings, the image of described two pre-sizings is pressed respectively to the pattern of row 2 × N+1 decile, row 2 × N+1 decile, be divided into (2 × N+1) × (2 × N+1) number of sub images, and the subimage of image to pre-sizing every described is numbered by its place ranks, described N is more than or equal to 1 integer.
3. method as claimed in claim 2, it is characterized in that, described from described subimage selected one or more subimages, for selected described one or more subimages, relatively itself and the pixel value of contiguous one or more subimages, and determine that according to comparative result coding comprises:
Step 1: from described subimage selected ((N+1), (N+1)) subimage, and by described ((N+1), (N+1)) subimage is as center subimage, wherein said ((N+1), (N+1)) is the row and column at subimage place;
Step 2: Jiang Gai center subimage subimage adjacent thereto carries out pixel value comparison, the pixel value of Ruo Gai center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
Step 3: subimage contiguous every number of sub images in Jiang Gai center is as center subimage, and return to step 2 and carry out, until the every number of sub images in the image of described pre-sizing is all completeer;
Step 4: delete the coding that repeats relatively to obtain, and arrange remaining coding by predefined procedure, coding after arranging is as the coding of described image to be identified, is wherein saidly remainingly encoded to the coding that step 2 and step 3 obtain and deducts the coding that repeats relatively to obtain.
4. method as claimed in claim 2, it is characterized in that, described from described subimage selected one or more subimages, for selected described one or more subimages, relatively itself and the pixel value of contiguous one or more subimages, and determine that according to comparative result coding comprises:
Step 1: from described subimage selected ((N+1), (N+1)) subimage, and by described ((N+1), (N+1)) subimage is as the first center subimage, wherein said ((N+1), (N+1)) is the row and column at subimage place;
Step 2: this first center subimage subimage adjacent thereto is carried out to pixel value comparison, if the pixel value of this first center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
Step 3: every number of sub images that this first center subimage is close to angular direction is as the second center subimage, and described the second center subimage subimage adjacent thereto is compared, if the pixel value of described the second center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
Step 4: using every number of sub images of this first center subimage vertical direction and horizontal direction vicinity as San center subimage, and the perpendicular direction of described San center subimage and the contiguous subimage of horizontal direction are compared, if the pixel value of center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
Step 5: using every number of sub images of every number of sub images contiguous described the second center subimage and San center subimage vertical direction and horizontal direction vicinity as the first center subimage, and return to execution step 2, until the every number of sub images in the image of described pre-sizing is all completeer;
Step 6: delete the coding that repeats relatively to obtain, and arrange remaining coding by predefined procedure, coding after arranging is as the coding of described image to be identified, is wherein saidly remainingly encoded to the coding that step 2 to step 5 obtains and deducts the coding that repeats relatively to obtain.
5. the method as described in claim 2 to 4 any one, is characterized in that, the height of the image of described two pre-sizings is identical,
The relational expression of width is W
4-3-get=(W
16-9-get× W
16-9)/W
4-3
Wherein, W
16-9, W
4-3for fixed value, W
16-9-getfor width, the W of first image intercepting
4-3-getfor the width of second image intercepting.
6. a pattern recognition device, is characterized in that, described device comprises:
Image acquisition unit, for obtaining the gray-value image of image to be identified;
Image division unit, for becoming multiple equal-sized subimages by described gray-value image by the mode division of row, column decile;
Coding unit, for selecting one or more subimages from described subimage, for selected described one or more subimages, relatively itself and the pixel value of contiguous one or more subimages, and determine coding according to comparative result, described coding and pre-stored Image Coding are mated to identify described image to be identified.
7. device as claimed in claim 6, is characterized in that, described image division unit specifically for:
From described gray-value image, intercept the image of two pre-sizings, the image of described two pre-sizings is pressed respectively to the pattern of row 2 × N+1 decile, row 2 × N+1 decile, be divided into (2 × N+1) × (2 × N+1) number of sub images, and the subimage of image to pre-sizing every described is numbered by its place ranks, described N is more than or equal to 1 integer.
8. device as claimed in claim 7, is characterized in that, described coding unit comprises:
The first coding module, for selecting ((N+1) from described subimage, (N+1)) subimage, and by described ((N+1), (N+1)) subimage is as center subimage, wherein said ((N+1), (N+1)) is the row and column at subimage place;
The second coding module, carries out pixel value comparison for Jiang Gai center subimage subimage adjacent thereto, the pixel value of Ruo Gai center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
The 3rd coding module, every number of sub images of being close to for Jiang Gai center subimage is as center subimage;
The first control module, carries out for the circulation of controlling the second coding module and the 3rd coding module, until the every number of sub images in the image of described pre-sizing is all completeer;
The 4th coding module, for deleting the coding that repeats relatively to obtain, and arrange remaining coding by predefined procedure, the coding after arranging is as the coding of described image to be identified, is wherein saidly remainingly encoded to the coding that the second coding module obtains and deducts the coding that repeats relatively to obtain.
9. device as claimed in claim 7, is characterized in that, described coding unit comprises:
The 5th coding module, for selecting ((N+1) from described subimage, (N+1)) subimage, and by described ((N+1), (N+1)) subimage is as the first center subimage, wherein said ((N+1), (N+1)) is the row and column at subimage place;
The 6th coding module, for this first center subimage subimage adjacent thereto is carried out to pixel value comparison, if the pixel value of this first center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
The 7th coding module, the every number of sub images that is used for this first center subimage to be close to angular direction is as the second center subimage, and described the second center subimage subimage adjacent thereto is compared, if the pixel value of described the second center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
The 8th coding module, be used for using every number of sub images of this first center subimage vertical direction and horizontal direction vicinity as San center subimage, and the perpendicular direction of described San center subimage and the contiguous subimage of horizontal direction are compared, if the pixel value of center subimage and be more than or equal to the pixel value of its subimage relatively and, be encoded to 1, otherwise be encoded to 0;
The 9th coding module, for using every number of sub images contiguous described the second center subimage and San center subimage vertical direction and the contiguous every number of sub images of horizontal direction as the first center subimage;
The second control module, carries out for the circulation of controlling the 6th coding module, the 7th coding module, the 8th coding module and the 9th coding module, until the every number of sub images in the image of described pre-sizing is all completeer;
The tenth coding module, for deleting the coding that repeats relatively to obtain, and arrange remaining coding by predefined procedure, coding after arranging is as the coding of described image to be identified, is wherein saidly remainingly encoded to the coding that the 6th coding module, the 7th coding module and the 8th coding module obtain and deducts the coding that repeats relatively to obtain.
10. the device as described in claim 7 to 9 any one, is characterized in that, the height of the image of described two pre-sizings is identical,
The relational expression of width is W
4-3-get=(W
16-9-get× W
16-9)/W
4-3
Wherein, W
16-9, W
4-3for fixed value, W
16-9-getfor width, the W of first image intercepting
4-3-getfor the width of second image intercepting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410179080.1A CN103942543B (en) | 2014-04-29 | 2014-04-29 | A kind of image-recognizing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410179080.1A CN103942543B (en) | 2014-04-29 | 2014-04-29 | A kind of image-recognizing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103942543A true CN103942543A (en) | 2014-07-23 |
| CN103942543B CN103942543B (en) | 2018-11-06 |
Family
ID=51190207
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410179080.1A Active CN103942543B (en) | 2014-04-29 | 2014-04-29 | A kind of image-recognizing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103942543B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104933670A (en) * | 2015-07-02 | 2015-09-23 | 北京电子工程总体研究所 | Storage method and system of image pixel data |
| CN105631821A (en) * | 2015-12-25 | 2016-06-01 | 北京像素软件科技股份有限公司 | Interface chromatic aberration diffusion method and device |
| CN108154031A (en) * | 2018-01-17 | 2018-06-12 | 腾讯科技（深圳）有限公司 | Recognition methods, device, storage medium and the electronic device of camouflage applications program |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101154265A (en) * | 2006-09-29 | 2008-04-02 | 中国科学院自动化研究所 | Method for recognizing iris with matched characteristic and graph based on partial bianry mode |
| US20080166026A1 (en) * | 2007-01-10 | 2008-07-10 | Samsung Electronics Co., Ltd. | Method and apparatus for generating face descriptor using extended local binary patterns, and method and apparatus for face recognition using extended local binary patterns |
| CN101567043A (en) * | 2009-05-31 | 2009-10-28 | 中山大学 | Face tracking method based on classification and identification |
| US20120070041A1 (en) * | 2010-09-16 | 2012-03-22 | Jie Wang | System And Method For Face Verification Using Video Sequence |
| CN103077378A (en) * | 2012-12-24 | 2013-05-01 | 西安电子科技大学 | Non-contact human face identifying algorithm based on expanded eight-domain local texture features and attendance system |
| CN103345516A (en) * | 2013-07-10 | 2013-10-09 | 百度在线网络技术（北京）有限公司 | Application program searching method based on application program icons and system |
-
2014
- 2014-04-29 CN CN201410179080.1A patent/CN103942543B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101154265A (en) * | 2006-09-29 | 2008-04-02 | 中国科学院自动化研究所 | Method for recognizing iris with matched characteristic and graph based on partial bianry mode |
| US20080166026A1 (en) * | 2007-01-10 | 2008-07-10 | Samsung Electronics Co., Ltd. | Method and apparatus for generating face descriptor using extended local binary patterns, and method and apparatus for face recognition using extended local binary patterns |
| CN101567043A (en) * | 2009-05-31 | 2009-10-28 | 中山大学 | Face tracking method based on classification and identification |
| US20120070041A1 (en) * | 2010-09-16 | 2012-03-22 | Jie Wang | System And Method For Face Verification Using Video Sequence |
| CN103077378A (en) * | 2012-12-24 | 2013-05-01 | 西安电子科技大学 | Non-contact human face identifying algorithm based on expanded eight-domain local texture features and attendance system |
| CN103345516A (en) * | 2013-07-10 | 2013-10-09 | 百度在线网络技术（北京）有限公司 | Application program searching method based on application program icons and system |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104933670A (en) * | 2015-07-02 | 2015-09-23 | 北京电子工程总体研究所 | Storage method and system of image pixel data |
| CN105631821A (en) * | 2015-12-25 | 2016-06-01 | 北京像素软件科技股份有限公司 | Interface chromatic aberration diffusion method and device |
| CN108154031A (en) * | 2018-01-17 | 2018-06-12 | 腾讯科技（深圳）有限公司 | Recognition methods, device, storage medium and the electronic device of camouflage applications program |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103942543B (en) | 2018-11-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102750564B (en) | Dynamic two-dimension code and coding/decoding method thereof | |
| US8879839B2 (en) | Image processing apparatus, image processing method, and storage medium | |
| CN112926715B (en) | Two-dimensional code generation and decoding method, device and equipment | |
| CN109146991B (en) | Picture format conversion method, device, equipment and storage medium | |
| KR20160130792A (en) | Method and apparatus for generating mosaic image | |
| CN104268615A (en) | Encoding method and decoding method for color two-dimensional code | |
| CN101930337A (en) | Method for processing on-screen display and associated embedded system | |
| CN105094289A (en) | Method, device, and system for realizing graphical user interface | |
| CN104035664A (en) | Display method and device for user interface of application program | |
| CN103929644A (en) | Video fingerprint database building method and device and video fingerprint recognition method and device | |
| CN103942543A (en) | Image recognition method and device | |
| CN109543770A (en) | Dot character recognition methods and device | |
| CN105069104A (en) | Dynamic cartoon generation method and system | |
| CN105654019A (en) | Fast decoding method and device for two-dimension code | |
| CN102075752A (en) | Image coding method, image decoding method and device for embedded system | |
| CN110399760A (en) | A kind of batch two dimensional code localization method, device, electronic equipment and storage medium | |
| CN105578035A (en) | Image processing method and electronic device | |
| CN109636862A (en) | Image processing method, system, terminal, storage medium and writing board device | |
| CN102306294A (en) | Method and system for extracting image from portable document format (PDF) file page | |
| CN103425980A (en) | Method and device for obtaining text regions in pictures | |
| CN110853063A (en) | Image segmentation information processing method, device, equipment and storage medium | |
| CN102542528B (en) | Image conversion processing method and system | |
| CN102495695B (en) | Multipoint touch identification method of infrared diode touch screen as well as touch device and system | |
| CN102663667B (en) | Storage method and display method for handwritten content on touch screen, terminal equipment | |
| CN110321759B (en) | Video feature extraction method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |