CN103034854A - Image processing device and image processing method - Google Patents
Image processing device and image processing method Download PDFInfo
- Publication number
- CN103034854A CN103034854A CN201110310159XA CN201110310159A CN103034854A CN 103034854 A CN103034854 A CN 103034854A CN 201110310159X A CN201110310159X A CN 201110310159XA CN 201110310159 A CN201110310159 A CN 201110310159A CN 103034854 A CN103034854 A CN 103034854A
- Authority
- CN
- China
- Prior art keywords
- binary map
- connected domain
- gray scale
- layer
- stroke
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
- Image Analysis (AREA)
Abstract
The invention provides an image processing device and an image processing method. The image processing device comprises a binary image generation division, a connected domain identification division and a stroke layer determination division. In the binary image generation division, binary images are generated corresponding to each layer of a ternary image based on the ternary image. In each binary image, pixels which belong to a corresponding layer are provided with first gray levels, and other pixels are provided with second gray levels. The connected domain identification division identifies a second gray level connected domain, which is not connected with a boundary, in each binary image. According to the stroke layer determination division, under the circumstances that the specific value of the gross area of the second gray level connected domains of the two binary images is greater than a first threshold, and the contact ratio between a first gray level area in the binary image with a smaller gross area of the second gray level connected domain and the second gray connected domain in the binary image with a bigger gross area of the second gray level connected domain is greater than a second threshold, the stroke layer determination division determines the layer, corresponding to the binary image with the smaller gross area of the second gray level connected domain, as a stroke layer and determines the layer, corresponding to the other binary image, as a stroke boundary area layer.
Description
Technical field
The present invention relates generally to image processing apparatus and method, be specifically related to for image processing apparatus and method from image identification stroke layer.
Background technology
Literal in the video is for the very succinct of video frequency searching (video indexing) and video frequency abstract (videosummarization) and clue accurately.Therefore, after the caption area in detecting video image, can carry out OCR (optical character identification) to obtain the Word message of captions to caption area.In this process, be very important step to the extraction of stroke.
One class conventional method is obtained stroke with the color cluster method.In these class methods, the color of supposing stroke is consistent.Yet for some video, this hypothesis also is false.In addition, for low image quality video image, colouring information is insecure sometimes.And, when background color and stroke color similarity, can introduce more noise.
Another kind of conventional method uses local binarization method (such as the Niblack algorithm etc.) to obtain stroke.Yet, because the characteristic of this algorithm also may be introduced more noise.
In addition, all can to run into which layer of how judging after processing image in the image be the problem of the stroke layer of reality to above-mentioned two kinds of methods.
In video image, especially in the caption area of video image, usually see that with the stroke of frontier district (for example, with the dark stroke of light frontier district or with the light stroke of dark frontier district) right and wrong this stroke frontier district is to can be used for the important information that stroke extracts.Yet existing stroke extracting method is also underused this feature.
Summary of the invention
The purpose of this invention is to provide a kind of apparatus and method of from image, identifying the stroke layer, to solve the aforementioned problems in the prior at least in part.
According to one embodiment of present invention, a kind of image processing apparatus is provided, comprise: the binary map generating portion, be configured to generate the binary map corresponding with every one deck of this three value figure according to three value figure, in each binary map, the pixel that belongs to equivalent layer has the first gray scale, and other pixel has the second gray scale; The connected domain identification division is configured to identify in each binary map not the second gray scale connected domain with boundary connected; And stroke layer determining section, be configured to ratio at the total area of the second gray scale connected domain of two binary map greater than first threshold, and in the situation of registration greater than Second Threshold between the second gray scale connected domain in the larger binary map of the first gray areas in the binary map that the second gray scale connected domain total area is less in these two binary map and the second gray scale connected domain total area, the corresponding layer of the binary map that the total area of the second gray scale connected domain in these two binary map is less is defined as the stroke layer, and the corresponding layer of another binary map is defined as stroke frontier district layer.
According to another embodiment of the invention, a kind of image processing method is provided, comprises step: figure generates the binary map corresponding with every one deck of this three value figure according to three values, in each binary map, the pixel that belongs to equivalent layer has the first gray scale, and other pixel has the second gray scale; Identify in each binary map not the second gray scale connected domain with boundary connected; And at the ratio of the total area of the second gray scale connected domain of two binary map greater than first threshold, and in the situation of registration greater than Second Threshold between the second gray scale connected domain in the larger binary map of the first gray areas in the binary map that the second gray scale connected domain total area is less in these two binary map and the second gray scale connected domain total area, the corresponding layer of the binary map that the total area of the second gray scale connected domain in these two binary map is less is defined as the stroke layer, and the corresponding layer of another binary map is defined as stroke frontier district layer.
Description of drawings
With reference to below in conjunction with the explanation of accompanying drawing to the embodiment of the invention, can understand more easily above and other purpose of the present invention, characteristics and advantage.For fear of having blured the present invention because of unnecessary details, only show in the accompanying drawings with according to the closely-related apparatus structure of the solution of the present invention and/or treatment step, and omitted other details little with relation of the present invention.
Fig. 1 is the block diagram that illustrates according to the ios dhcp sample configuration IOS DHCP of the image processing apparatus of the embodiment of the invention;
Fig. 2 is the example that the image that comprises stroke object and stroke frontier district is shown;
Fig. 3 a to Fig. 3 c illustrates the binary map corresponding with the every one deck among the three value figure that generate according to the image of Fig. 2;
Fig. 4 a to Fig. 4 c illustrates the hole connected domain in the binary map of Fig. 3 a to Fig. 3 c;
The binary map that produces in the processing of the way of example that the hole connected domain that obtains in the binary map of illustrating Fig. 5 a to Fig. 5 c adopts;
Fig. 6 illustrates the according to another embodiment of the present invention block diagram of the ios dhcp sample configuration IOS DHCP of image processing apparatus;
Fig. 7 illustrates the outline of each connected domain in the binary map shown in Fig. 3 b;
Fig. 8 a and Fig. 8 b illustrate respectively a stroke object connected domain among Fig. 3 b and the outline of this connected domain;
Fig. 9 a and Fig. 9 b illustrate respectively the connected domain of the non-stroke object (noise) among Fig. 3 b and the outline of this connected domain;
Figure 10 illustrates the example image of the stroke layer behind noise remove;
Figure 11 illustrates the process flow diagram according to the image processing method of the embodiment of the invention;
Figure 12 is the block diagram that the example arrangement of the computing machine of realizing equipment of the present invention and method is shown.
Embodiment
Embodiments of the invention are described with reference to the accompanying drawings.Should be noted that for purpose clearly, omitted expression and the description of parts that have nothing to do with the present invention, known to persons of ordinary skill in the art and processing in accompanying drawing and the explanation.
Fig. 1 is the block diagram that illustrates according to the configuration of the image processing apparatus 100 of the embodiment of the invention.Image processing apparatus 100 comprises binary map generating portion 110, connected domain identification division 120 and stroke layer determining section 130.
Image processing apparatus 100 be input as three value figure, this three value figure can utilize usual manner that image is carried out three values and obtain.For example, Fig. 2 illustrates the example image of the caption area in the video image, and in image shown in Figure 2, stroke is black, and background (video pictures) has gradient color, and has the frontier district of white between stroke and the background.Can adopt existing mode that this image is processed to generate three value figure, for example, can be with this image transitions gray level image (for example, the gray-scale value scope is 0 to 255), and by the gray scale of each pixel and two local window threshold values are compared and with image three values (that is, image being divided into three layers).As a specific example, can be T with the local window threshold value setting
1=m-k * s, T
2=m+k * s, wherein, m is local (for example 3 * 3 pixel windows) average, and s is local variance, and coefficient k can be set to different values as required.By comparing with threshold value, can be with gray-scale value less than T
1The set of pixels cooperation be black layer, with gray-scale value greater than T
2The set of pixels cooperation be white layer, and with gray-scale value between T
1And T
2Between the set of pixels cooperation be the middle layer.In addition, those skilled in the art can expect, also can adopt the alternate manner such as color cluster etc. to obtain three value figure.
Binary map generating portion 110 generates the binary map corresponding with every one deck of this three value figure according to the three value figure that input, and namely extracts every one deck of three value figure.In each binary map, the pixel that belongs to equivalent layer has the first gray scale, and other pixel (background) has the second gray scale.
Fig. 3 a to Fig. 3 c illustrates three binary map that generated according to three value figure of the image of Fig. 2 by binary map generating portion 110.In the binary map of Fig. 3 a to Fig. 3 c, the pixel that belongs to the equivalent layer of three value figure is black, and other pixel (background) is white.
Binary map generating portion 110 offers connected domain identification division 120 with the binary map that generates.In each binary map of connected domain identification division 120 identification not with the second gray scale connected domain of boundary connected.Hereinafter, with not with the second gray scale connected domain of boundary connected referred to as the hole connected domain.
Fig. 4 a to Fig. 4 c illustrates the hole connected domain that is identified by connected domain identification division 120 from the binary map of Fig. 3 a to Fig. 3 c.
As a concrete example, connected domain identification division 120 can obtain hole connected domain in the binary map by following processing:
Binary map is carried out inverse to be processed;
The first gray scale connected domain with boundary connected in the binary map after inverse processed is converted to the second gray scale;
With the conversion after binary map in the first gray areas as the hole connected domain.
Fig. 5 a to Fig. 5 c illustrates the example of each binary map in this example process.Fig. 5 a is the binary map of the object of this processing, the binary map of Fig. 5 b for the binary map of Fig. 5 a being carried out obtain after inverse is processed, and Fig. 5 c is for being converted to the binary map that obtains after the second gray scale with the connected domain of boundary connected among Fig. 5 b.
Yet those skilled in the art can expect, can adopt other connected domain analysis method to identify hole connected domain in the binary map.
Connected domain identification division 120 will offer for the connected domain recognition result that each binary map is carried out stroke layer determining section 130.Stroke layer determining section 130 determines whether to exist in three binary map binary map corresponding to stroke layer and stroke frontier district layer according to the comparative result of the hole connected domain in each binary map, and which binary map is corresponding to the stroke layer, and which binary map is corresponding to stroke frontier district layer.
Particularly, the ratio of the total area (being pixel quantity) of the hole connected domain of stroke layer determining section 130 in can more any two binary map, when the total area ratio of the hole connected domain of two binary map surpassed predetermined threshold (for example 20), stroke layer determining section 130 can judge that the corresponding layer of these two binary map may be respectively stroke layer and stroke frontier district.This is because the total area of the hole connected domain in the corresponding binary map of stroke frontier district layer is usually larger; and the total area of the hole connected domain in the corresponding binary map of stroke layer is usually less, so the ratio of their the hole connected domain total area usually can be high significantly.
In addition, hole connected domain total area ratio two binary map surpasses in the situation of predetermined threshold, stroke layer determining section 130 can further compare the registration between the hole connected domain in the larger binary map of the first gray areas in the less binary map of these two binary map Hole connected domain total areas and the hole connected domain total area, in the situation of this registration greater than predetermined threshold, stroke layer determining section 130 can determine that less corresponding layer of these two binary map Hole connected domain total areas is the stroke layer, and the corresponding layer of another one is stroke frontier district layer.
Take each binary map shown in Figure 3 as example, stroke layer determining section 130 can compare hole connected domain recognition result shown in Figure 4, at the ratio of the total area of the hole connected domain of determining Fig. 4 a and the total area of the hole connected domain shown in Fig. 4 b above predetermined threshold, and the registration of black region surpasses in the situation of predetermined threshold among the hole connected domain shown in Fig. 4 a and Fig. 3 b, can determine that the corresponding layer of Fig. 3 b is the stroke layer, and the corresponding layer of Fig. 3 a is stroke frontier district layer.
According to a specific embodiment of the present invention, can be with N1/N2 as above-mentioned registration, wherein, N1 is the pixel quantity of the common factor of the hole connected domain in the larger binary map of the total area of the first gray areas in the less binary map of the total area in the logical territory of hole and hole connected domain; N2 is the pixel quantity of the union of the hole connected domain in the larger binary map of the first gray areas in the less binary map of the hole connected domain total area and the hole connected domain total area.In this case, the threshold value of above-mentioned registration for example can be selected in from 0.6 to 0.9 the scope, and for example 0.7.Yet those skilled in the art can expect also can measuring above-mentioned registration in other concrete mode.
In addition, under some individual cases, the hole connected domain total area in certain binary map very little (for example less than this binary map total area 1%), it is disconnected to make the erroneous judgement that has stroke layer and stroke frontier district layer.Therefore, in one embodiment, except above-mentioned two Rule of judgment, stroke layer determining section 130 can also additionally adopt following Rule of judgment to improve the accuracy of judging:
For these two binary map each, the ratio of the area of hole connected domain and the total area of this binary map is higher than predetermined threshold (for example, 0.05).
Image processing apparatus 100 can be exported determined stroke layer and/or stroke frontier district layer as result, being used for subsequent treatment, such as literal identification etc.Thereby, compare with the normal image treating apparatus, image processing apparatus is by utilizing the feature of the stroke frontier district in the image according to an embodiment of the invention, can identify efficiently and definite image in stroke layer and stroke frontier district layer.
In addition, according to another embodiment, image processing apparatus can also carry out denoising to determined stroke layer.
Fig. 6 is the block diagram that the ios dhcp sample configuration IOS DHCP of image processing apparatus 600 according to another embodiment of the invention is shown.Image processing apparatus 600 comprises binary map generating portion 610, connected domain identification division 620, stroke layer determining section 630 and noise remove part 640.Wherein, the configuration of binary map generating portion 610, connected domain identification division 620 and stroke layer determining section 630 is omitted detailed description thereof with similar with reference to binary map generating portion 110, connected domain identification division 120 and the stroke layer determining section 130 of Fig. 1 explanation at this.
Noise remove part 640 is set to the stroke frontier district layer of determining according to by stroke layer determining section 630, and determined stroke layer is carried out denoising.Particularly, noise remove part 640 is for each the first gray scale connected domain in the binary map that is confirmed as the stroke layer, in the goodness of fit the situation less than predetermined threshold of outline with respect to the first gray areas in the binary map of determined stroke frontier district layer of this first gray scale connected domain, with this first gray scale connected domain as noise remove.Usually understand such as those skilled in the art, the outline of connected domain refers to the profile of the border adjacency of and this connected domain outside in this connected domain.
Still take binary map shown in Figure 3 as example, in stroke layer determining section 630 the corresponding layer of the binary map of Fig. 3 b is defined as the stroke layer, and the corresponding layer of the binary map of Fig. 3 a is defined as in the situation of stroke frontier district layer, noise remove part 640 can be extracted the outline of each the first gray scale connected domain in the binary map of stroke layer.Fig. 7 shows the outline of each connected domain in the binary map shown in Fig. 3 b.In each the first gray scale connected domain of the binary map of stroke layer, may comprise real stroke object, for example shown in Fig. 8 a, also may comprise noise, shown in Fig. 8 b.Noise remove part 640 is by comparing the outline (for example, shown in Fig. 8 b and Fig. 9 b) of this connected domain and the first gray areas in the layer of stroke frontier district.Can find out, connected domain outline shown in Fig. 8 b can with Fig. 3 a in the first gray areas coincide, therefore noise remove part 640 can determine that the connected domain shown in Fig. 8 a is corresponding to the stroke object, yet the connected domain outline shown in Fig. 9 b can not with Fig. 3 a in the first gray areas coincide, so noise remove part 640 can be with the connected domain of Fig. 9 a as noise remove.Figure 10 illustrates the example image of the stroke layer after noise remove part 640 is removed noise.
According to a specific embodiment, can be with N4/N3 as the above-mentioned goodness of fit, wherein, N3 is the pixel quantity of the outline of the first gray scale connected domain in the binary map of stroke layer; N4 is the pixel quantity of the common factor of the first gray areas in the binary map of the outline of this first gray scale connected domain and stroke frontier district layer.In this case, the threshold value of the above-mentioned goodness of fit for example can be selected in from 0.85 to 0.95 the scope, and for example 0.9.Yet those skilled in the art can expect also can measuring the above-mentioned goodness of fit in other concrete mode.
Figure 11 illustrates the process flow diagram according to the image processing method of the embodiment of the invention.
At step S1110, generate the binary map corresponding with every one deck of this three value figure according to three value figure, in each binary map, the pixel that belongs to equivalent layer has the first gray scale, and other pixel has the second gray scale;
At step S1120, identify in each binary map not the second gray scale connected domain with boundary connected;
At step S1130, determine two binary map not with the ratio of the total area of the second gray scale connected domain of boundary connected whether greater than first threshold (for example, 20), if determine that the result is yes, process proceeds to step S1140;
At step S1140, determine in these two binary map not and binary map that the total area of the second gray scale connected domain of boundary connected is less in the first gray areas and or not in the larger binary map of the total area of the second gray scale connected domain of boundary connected not and the registration between the second gray scale connected domain of boundary connected whether greater than Second Threshold, if determine that the result is yes, then process proceeds to step S1150;
At step S1150, will not be defined as the stroke layer with the less corresponding layer of binary map of the total area of the second gray scale connected domain of boundary connected in these two binary map, and the corresponding layer of another binary map is defined as stroke frontier district layer.
According to an embodiment, in step S1140, with N1/N2 as described registration, wherein, N1 be not with the less binary map of the total area of the second gray scale connected domain of boundary connected in the first gray areas with or not the larger binary map of the total area of the second gray scale connected domain of boundary connected in not with the pixel quantity of the common factor of the second gray scale connected domain of boundary connected; N2 be not with the less binary map of the total area of the second gray scale connected domain of boundary connected in the first gray areas with or not the larger binary map of the total area of the second gray scale connected domain of boundary connected in not with the pixel quantity of the union of the second gray scale connected domain of boundary connected.
According to an embodiment, after step S1150, also comprise the noise remove step:
For each the first gray scale connected domain in the binary map of determined stroke layer, in the goodness of fit the situation less than three threshold value of outline with respect to the first gray areas in the binary map of determined stroke frontier district layer of this first gray scale connected domain, with this first gray scale connected domain as noise remove.
According to an embodiment, in the noise remove step, as the described goodness of fit, wherein, N3 is the pixel quantity of the outline of this first gray scale connected domain in the binary map of stroke layer with N4/N3; N4 is the pixel quantity of the common factor of the first gray areas in the binary map of the outline of this first gray scale connected domain in the binary map of stroke layer and stroke frontier district layer.
According to an embodiment, definite result at step S1140 is in the situation that is, process proceeds to step: determine in these two binary map each not with the ratio of the total area of the total area of the second gray scale connected domain of boundary connected and this binary map whether greater than the 4th threshold value (for example 0.05), be in the situation that is in definite result, process proceeds to step S1150.
The person of ordinary skill in the field knows that the present invention can be presented as device, method or computer program.Therefore, the present invention can specific implementation be following form, that is, can be completely hardware, the completely combination of software (comprising firmware, resident software, microcode etc.) or software section and hardware components.In addition, the present invention can also take to be embodied in the form of the computer program in any tangible expression medium, comprises the procedure code that computing machine can be used in this medium.
Can use any combination of one or more computer-readable mediums.Computer-readable medium can be computer-readable signal media or computer-readable recording medium, computer-readable recording medium for example can be, but be not limited to, electricity, magnetic, light, electromagnetism, ultrared or semi-conductive system, device, device or propagation medium or aforementioned every any suitable combination.The more specifically example of computer-readable recording medium (non exhaustive tabulation) comprising: electrical connection, portable computer diskette, hard disk, random access memory (RAM), ROM (read-only memory) (ROM), erasable type programmable read only memory (EPROM or flash memory), optical fiber, Portable, compact disk ROM (read-only memory) (CD-ROM), light storage device, magnetic memory device or aforementioned every any suitable combination of one or more wires are arranged.In this paper linguistic context, computer-readable recording medium can be anyly to contain or store for tangible medium instruction execution system, device or device or the program that and instruction executive system, device or device interrelate.
Be used for carrying out the computer program code of operation of the present invention, can write with any combination of one or more programming languages, described programming language comprises object oriented program language-such as Java, Smalltalk, C++, also comprise conventional process type programming language-such as " C " programming language or similar programming language.Procedure code can fully be carried out at user's computing machine, partly carries out at user's computing machine, carry out or carry out at remote computer or server fully at remote computer as part on an independently software package execution, the computing machine of part the user.In rear a kind of situation, remote computer can be by any kind network-comprise LAN (Local Area Network) (LAN) or wide area network (WAN)-be connected to user's computing machine, perhaps, can (for example utilize the ISP to pass through the Internet) and be connected to outer computer.
Figure 12 is the block diagram that the example arrangement of the computing machine of realizing equipment of the present invention and method is shown.In Figure 12, CPU (central processing unit) (CPU) 1201 carries out various processing according to the program of storage in the ROM (read-only memory) (ROM) 1202 or from the program that storage area 1208 is loaded into random access memory (RAM) 1203.In RAM 1203, also store as required data required when CPU 1201 carries out various processing etc.
Following parts are connected to input/output interface 1205: importation 1206 comprises keyboard, mouse etc.; Output 1207 comprises display, such as cathode-ray tube (CRT) (CRT), liquid crystal display (LCD) etc., and loudspeaker etc.; Storage area 1208 comprises hard disk etc.; With communications portion 1209, comprise that network interface unit is such as LAN card, modulator-demodular unit etc.Communications portion 1209 is processed such as the Internet executive communication via network.
As required, driver 1210 also is connected to input/output interface 1205.Detachable media 1211 is installed on the driver 1210 as required such as disk, CD, magneto-optic disk, semiconductor memory etc., so that the computer program of therefrom reading is installed in the storage area 1208 as required.
Realizing by software in the situation of above-mentioned steps and processing, such as detachable media 1211 program that consists of software is being installed such as the Internet or storage medium from network.
It will be understood by those of skill in the art that this storage medium is not limited to shown in Figure 12 wherein has program stored therein, distributes separately to provide the detachable media 1211 of program to the user with method.The example of detachable media 1211 comprises disk, CD (comprising compact disc read-only memory (CD-ROM) and digital universal disc (DVD)), magneto-optic disk (comprising mini-disk (MD)) and semiconductor memory.Perhaps, storage medium can be hard disk that comprises in ROM 1202, the storage area 1208 etc., computer program stored wherein, and be distributed to the user with the method that comprises them.
The device of the counter structure in the claim, operation and all functions restriction or step be equal to replacement, be intended to comprise any for carry out structure or the operation of this function with other unit of specifically noting in the claims combinedly.Its purpose of the given description of this invention is signal and describes, and is not to be exhaustive, also is not to be to be limited to the form of explaining to the present invention.For the person of an ordinary skill in the technical field, in the situation that does not depart from the scope of the invention and spirit, obviously can make many modifications and modification.To selection and the explanation of embodiment, be in order to explain best principle of the present invention and practical application, the person of an ordinary skill in the technical field can be understood that the present invention can have the various embodiments with various changes that are fit to desired special-purpose.
Claims (10)
1. image processing apparatus comprises:
The binary map generating portion is configured to generate the binary map corresponding with every one deck of described three value figure according to three value figure, and in each described binary map, the pixel that belongs to equivalent layer has the first gray scale, and other pixel has the second gray scale;
The connected domain identification division is configured to identify in each described binary map not the second gray scale connected domain with boundary connected; And
Stroke layer determining section, be configured to ratio at the total area of the described second gray scale connected domain of two described binary map greater than first threshold, and in the situation of registration greater than Second Threshold between described the second gray scale connected domain in the larger binary map of the first gray areas in the binary map that the total area of the second gray scale connected domain is less described in described two binary map and the total area of described the second gray scale connected domain, the corresponding layer of the binary map that the total area of the second gray scale connected domain described in described two binary map is less is defined as the stroke layer, and the corresponding layer of another binary map is defined as stroke frontier district layer.
2. device according to claim 1, wherein with N1/N2 as described registration, wherein,
N1 is the pixel quantity of the common factor of described the second gray scale connected domain in the larger binary map of the total area of the first gray areas in the less binary map of the total area of described the second gray scale connected domain and described the second gray scale connected domain;
N2 is the pixel quantity of the union of described the second gray scale connected domain in the larger binary map of the total area of the first gray areas in the less binary map of the total area of described the second gray scale connected domain and described the second gray scale connected domain.
3. device as claimed in claim 1 or 2 also comprises:
The noise remove part, be configured to for each the first gray scale connected domain in the binary map of determined stroke layer, in the goodness of fit the situation less than three threshold value of outline with respect to the first gray areas in the binary map of determined stroke frontier district layer of described the first gray scale connected domain, with this first gray scale connected domain as noise remove.
4. device as claimed in claim 3, wherein with N4/N3 as the described goodness of fit, wherein,
N3 is the pixel quantity of the outline of described the first gray scale connected domain in the binary map of described stroke layer;
N4 is the pixel quantity of the common factor of the first gray areas in the binary map of the outline of described the first gray scale connected domain in the binary map of described stroke layer and described stroke frontier district layer.
5. device as claimed in claim 1 or 2, wherein, described stroke layer determining section to described stroke layer and described stroke frontier district layer determine also need satisfy condition:
The ratio of the total area of the described second gray scale connected domain of each in described two binary map and the total area of this binary map is greater than the 4th threshold value.
6. image processing method comprises step:
Generate the binary map corresponding with every one deck of described three value figure according to three value figure, in each described binary map, the pixel that belongs to equivalent layer has the first gray scale, and other pixel has the second gray scale;
Identify in each described binary map not the second gray scale connected domain with boundary connected; And
At the ratio of the total area of the described second gray scale connected domain of two described binary map greater than first threshold, and in the situation of registration greater than Second Threshold between described the second gray scale connected domain in the larger binary map of the first gray areas in the binary map that the total area of the second gray scale connected domain is less described in described two binary map and the total area of described the second gray scale connected domain, the corresponding layer of the binary map that the total area of the second gray scale connected domain described in described two binary map is less is defined as the stroke layer, and the corresponding layer of another binary map is defined as stroke frontier district layer.
7. method according to claim 6, wherein with N1/N2 as described registration, wherein,
N1 is the pixel quantity of the common factor of described the second gray scale connected domain in the larger binary map of the total area of the first gray areas in the less binary map of the total area of described the second gray scale connected domain and described the second gray scale connected domain;
N2 is the pixel quantity of the union of described the second gray scale connected domain in the larger binary map of the total area of the first gray areas in the less binary map of the total area of described the second gray scale connected domain and described the second gray scale connected domain.
8. such as claim 6 or 7 described methods, also comprise step:
For each the first gray scale connected domain in the binary map of determined stroke layer, in the goodness of fit the situation less than three threshold value of outline with respect to the first gray areas in the binary map of determined stroke frontier district layer of described the first gray scale connected domain, with this first gray scale connected domain as noise remove.
9. method as claimed in claim 8, wherein with N4/N3 as the described goodness of fit, wherein,
N3 is the pixel quantity of the outline of described the first gray scale connected domain in the binary map of described stroke layer;
N4 is the pixel quantity of the common factor of the first gray areas in the binary map of the outline of described the first gray scale connected domain in the binary map of described stroke layer and described stroke frontier district layer.
10. such as claim 6 or 7 described methods, wherein, also need satisfy condition to the definite of described stroke layer and described stroke frontier district layer:
The ratio of the total area of the described second gray scale connected domain of each in described two binary map and the total area of this binary map is greater than the 4th threshold value.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110310159.XA CN103034854B (en) | 2011-09-30 | 2011-09-30 | Image processing apparatus and method |
| JP2012216955A JP6011213B2 (en) | 2011-09-30 | 2012-09-28 | Image processing apparatus and image processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110310159.XA CN103034854B (en) | 2011-09-30 | 2011-09-30 | Image processing apparatus and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103034854A true CN103034854A (en) | 2013-04-10 |
| CN103034854B CN103034854B (en) | 2016-03-30 |
Family
ID=48021733
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110310159.XA Active CN103034854B (en) | 2011-09-30 | 2011-09-30 | Image processing apparatus and method |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP6011213B2 (en) |
| CN (1) | CN103034854B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104254022A (en) * | 2014-09-25 | 2014-12-31 | 小米科技有限责任公司 | Subtitle displaying method and device |
| CN106845473A (en) * | 2015-12-03 | 2017-06-13 | 富士通株式会社 | For determine image whether be the image with address information method and apparatus |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108596874B (en) * | 2018-03-17 | 2024-01-05 | 紫光汇智信息技术有限公司 | Image definition judging method and device, computer equipment and product |
| CN112686264B (en) * | 2020-12-31 | 2023-10-20 | 中广核研究院有限公司 | Digital meter reading method, device, computer equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593276A (en) * | 2008-05-29 | 2009-12-02 | 汉王科技股份有限公司 | A kind of video OCR image-text separation method and system |
| US20100202690A1 (en) * | 2009-02-12 | 2010-08-12 | Samsung Electronics Co., Ltd. | Method for recognizing text from image |
| CN102054271A (en) * | 2009-11-02 | 2011-05-11 | 富士通株式会社 | Text line detection method and device |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4861845B2 (en) * | 2007-02-05 | 2012-01-25 | 富士通株式会社 | Telop character extraction program, recording medium, method and apparatus |
| JP2009017325A (en) * | 2007-07-06 | 2009-01-22 | Hitachi Ltd | Telop character region extraction device and method |
| JP2009217303A (en) * | 2008-03-07 | 2009-09-24 | Hitachi Ltd | Telop character extraction method and telop character recognition device |
| JP2010219858A (en) * | 2009-03-17 | 2010-09-30 | Seiko Epson Corp | Noise component removing device, and medium having noise component removing program recorded thereon |
-
2011
- 2011-09-30 CN CN201110310159.XA patent/CN103034854B/en active Active
-
2012
- 2012-09-28 JP JP2012216955A patent/JP6011213B2/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593276A (en) * | 2008-05-29 | 2009-12-02 | 汉王科技股份有限公司 | A kind of video OCR image-text separation method and system |
| US20100202690A1 (en) * | 2009-02-12 | 2010-08-12 | Samsung Electronics Co., Ltd. | Method for recognizing text from image |
| CN102054271A (en) * | 2009-11-02 | 2011-05-11 | 富士通株式会社 | Text line detection method and device |
Non-Patent Citations (1)
| Title |
|---|
| 邹星: "车牌识别中的图像提取和分割算法", 《重庆工学院学报(自然科学)》, vol. 23, no. 8, 31 August 2009 (2009-08-31), pages 19 - 28 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104254022A (en) * | 2014-09-25 | 2014-12-31 | 小米科技有限责任公司 | Subtitle displaying method and device |
| CN104254022B (en) * | 2014-09-25 | 2018-10-12 | 小米科技有限责任公司 | Caption presentation method and device |
| CN106845473A (en) * | 2015-12-03 | 2017-06-13 | 富士通株式会社 | For determine image whether be the image with address information method and apparatus |
| CN106845473B (en) * | 2015-12-03 | 2020-06-02 | 富士通株式会社 | Method and device for determining whether image is image with address information |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2013081174A (en) | 2013-05-02 |
| JP6011213B2 (en) | 2016-10-19 |
| CN103034854B (en) | 2016-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9619735B1 (en) | Pure convolutional neural network localization | |
| JP5775225B2 (en) | Text detection using multi-layer connected components with histograms | |
| CN109918513B (en) | Image processing method, device, server and storage medium | |
| US7471826B1 (en) | Character segmentation by slices | |
| CN102890780B (en) | Image processing device and image processing method | |
| US10318883B2 (en) | Identifying optimum times at which to retrain a logistic regression model | |
| CN113657274B (en) | Table generation method and device, electronic equipment and storage medium | |
| CN103383732A (en) | Image processing method and device | |
| CN108170806B (en) | Sensitive word detection and filtering method and device and computer equipment | |
| CN108734161B (en) | Method, device and equipment for identifying prefix number area and storage medium | |
| CN112200193B (en) | Distributed license plate recognition method, system and device based on multi-attribute fusion | |
| CN103034854B (en) | Image processing apparatus and method | |
| CN112329617B (en) | New scene face recognition model construction method and system based on single source domain sample | |
| CN110737770B (en) | Text data sensitivity identification method and device, electronic equipment and storage medium | |
| CN113221906A (en) | Image sensitive character detection method and device based on deep learning | |
| US9773472B2 (en) | Text extraction from graphical user interface content | |
| JP6303671B2 (en) | Image processing apparatus and image processing program | |
| CN112801960B (en) | Image processing method and device, storage medium and electronic equipment | |
| CN114170604A (en) | Character recognition method and system based on Internet of things | |
| JP2012222581A (en) | Image processing device, image processing method, program, and storage medium | |
| US20210081662A1 (en) | Scalable structure learning via context-free recursive document decomposition | |
| JP2017010187A (en) | Image processing device and image processing program | |
| CN117523644B (en) | Public transportation identity authentication method and system | |
| US11269626B2 (en) | Quality analysis of source code | |
| CN113450351B (en) | Segmentation model training method, image segmentation method, device, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |