CN105744260A - Bar pattern detection method and device - Google Patents
Bar pattern detection method and device Download PDFInfo
- Publication number
- CN105744260A CN105744260A CN201410745508.4A CN201410745508A CN105744260A CN 105744260 A CN105744260 A CN 105744260A CN 201410745508 A CN201410745508 A CN 201410745508A CN 105744260 A CN105744260 A CN 105744260A
- Authority
- CN
- China
- Prior art keywords
- pattern
- bar
- module
- bar paten
- difference
- 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
- Image Analysis (AREA)
Abstract
The invention discloses a bar pattern detection method and device. The method comprises the steps of: converting an image format of a pattern to be detected into a YUV format; horizontally scanning a converted pattern to be detected to search for boundary points; setting a target width of a bar area to be a horizontal coordinate spacing between adjacent boundary points; calculating the difference value one by one and line by line between the horizontal coordinate spacing of adjacent boundary points and the target width, calculating whether the difference value of each line is within a preset range, determining the line to be consistent with bar pattern characteristics if yes; determining the detected pattern to be a bar pattern if the number of lines exceeding a preset proportion is consistent with the bar pattern characteristics; or determining the detected pattern to be a non-bar pattern. According to the invention, the method does not select a fixed target width, can be used for detecting the bar pattern of any order, and detecting all scanning lines of each frame of image, and improves the accuracy of determining bar patterns.
Description
Technical field
The present invention relates to technical field of image detection, particularly relate to a kind of bar paten detection method and device.
Background technology
At present in technical field of image processing, the application of the bar shaped test pattern such as colour bar and GTG is extensive in the extreme.In the process relating to processing television signals, for instance being fabricated into transmitting transmission and receiving from TV programme, arrive including in the production or maintenance of television equipment, the pattern such as colour bar and GTG is all commonly used.Additionally, be applied to CVBS(CompositeVideoBroadcastSignal at DVD player, video camera, video doorbell monitoring, vehicle-mounted Rear View Display etc., Composite Video Baseband Signal) signal scene in, colour bar and gray scale image are commonly used especially.
By the colour bar inputted and gray scale image, it is possible to more convenient, understand the ruuning situation of equipment intuitively, engineering staff can carry out color correction according to colour bar pattern, and vedio color correction accurately is extremely important, especially home theater scialyscope etc..Because using different projection screens will obtain different color effect in different environments, only carry out the color correction guarantee display device display trichromatic accuracy of RGB.If additionally, not GTG performance accurately, serious heterochromia problem, white balance distortion will be exposed on multi-level color transition.So, on the basis of color correction, it is also desirable to GTG excessively just can present the most real color representation accurately, it is seen that colour bar and grey scale pattern are particularly significant in engineer applied.Therefore, in order to be beneficial to understanding system equipment situation and input video source carry out color correction and gray level white balance correction, the necessary video image that colour bar or GTG are inputted detects.
In prior art, the method for video color bar detection is that signal colour is analyzed, but the difference due to each input signal source, especially after the conversion of repeatedly encoding and decoding, indicator difference in colour bar signal colourity is obvious, the method of this direct analysis carrier chrominance signal is difficult to carry out colour bar detection accurately, reduces the accuracy rate of detection.
And another kind method judges the boundary point colour bar feature whether conformance with standard is preset according to the position of colour bar pattern colors rank boundary point, so that it is determined that whether the picture frame of detection is colour bar pattern.The method does not consider further that the chrominance information of colour bar, avoid the colourity difference caused due to different video input source and cause erroneous judgement, but the method can only the colour bar pattern of the fixing exponent number of criterion, after the colour bar width difference of input, the method equally also can not judge colour bar pattern.The colour bar pattern of fixing exponent number refers to the colour bar pattern of the green purplish red blue-black of white HUANGQING that standard signal source produces, the actually one of the more than standard of colour bar pattern, it is also possible to exponent number is less or more.Especially grey scale pattern, exponent number variable in reality is more.
Summary of the invention
The technical problem to be solved is to provide a kind of bar paten detection method and device, and it can detect the bar image of any exponent number.
For solving the technical problem of the present invention, disclosure one bar paten detection method, comprise the following steps:
It is yuv format by the image format conversion of tested pattern；
Tested pattern after horizontal sweep conversion, searches for boundary point；
Set the target width horizontal coordinate spacing as adjacent two boundary points of bar-shaped zone；
Calculating the horizontal coordinate spacing of adjacent two boundary points and the difference of described target width line by line one by one, whether the described difference of statistics each row is in preset range, and if so, then this row meets bar paten feature；If the line number exceeding preset ratio meets bar paten feature, then described tested pattern is bar paten；Otherwise, tested pattern is non-bar paten.
For solving the technical problem of the present invention, invention additionally discloses a kind of bar paten detecting device, determine module, width calculation module and judge module including format converting module, search module, target width；
It is yuv format that described format converting module is used for the image format conversion of tested pattern；
Described search module, for the tested pattern after horizontal sweep conversion, searches for boundary point；
Described target width determines that module for being set to the target width of bar-shaped zone by the horizontal coordinate spacing of adjacent two boundary points；
Described width calculation module is for calculating the horizontal coordinate spacing of adjacent two boundary points and the difference of described target width line by line one by one, and whether the described difference of statistics each row is in preset range, and if so, then this row meets bar paten feature；
Described judge module for judging whether that the line number exceeding preset ratio meets bar paten feature, be if so, then described tested pattern is bar paten；Otherwise, tested pattern is non-bar paten.
Compared with prior art, there is advantages that the present invention is by converting tested pattern to yuv format, choose boundary point, calculate target width, the horizontal coordinate spacing of every pair of adjacent boundary point of every for tested pattern a line and target width are contrasted, such that it is able to judge whether tested pattern is bar paten, visible, the target width of the present invention is not fixing, may be used for the bar paten of detection arbitrary order, and all scanning provisional capitals of a two field picture are detected, improves the accuracy judging bar paten.
Accompanying drawing explanation
Fig. 1 is the bar paten detection method block diagram of one embodiment of the invention；
Fig. 2 is the bar paten detection method block diagram of another embodiment of the present invention；
Fig. 3 is the bar paten structure of the detecting device figure of one embodiment of the invention；
Fig. 4 is the bar paten structure of the detecting device figure of another embodiment of the present invention.
Detailed description of the invention
Below in conjunction with drawings and Examples, the present invention is described in further detail.
As it is shown in figure 1, be the bar paten detection method block diagram of one embodiment of the invention, the detection method of the present embodiment comprises the following steps:
S100: be yuv format by the image format conversion of tested pattern；
Specifically, because the tested pattern source form of input is different, it is necessary to be converted into yuv format by unified for the video image of various forms, to obtain brightness Y and chrominance C, it is simple to the process of subsequent step.Wherein yuv format includes YUV420, YUV422 etc..
S200: the tested pattern after horizontal sweep conversion, searches for boundary point；
Specifically, this step can realize according to boundary search algorithm, can being such as luminance detail value searching method, the detail section in a frame testing image be judged by luminance detail value searching algorithm application searches algorithm on luminance component Y, and namely the obvious region of details is regarded as the region that border exists.Such as, the region having vertical bar is detail areas, and solid color regions is smooth area.In the luminance detail value searching method of this step, tested pattern after horizontal sweep conversion, judge whether each pixel is boundary point, by detecting the brightness value sum of two pixels in the left side of current pixel point, difference with the brightness value sum of two pixels in the right, it is judged that whether this difference judges more than default luminance threshold, if, then this pixel is boundary point, and otherwise this pixel is large area smooth region.Wherein luminance threshold dynamically can be configured by systems soft ware.
The specific embodiment of the invention is not limited to above-mentioned border search method, for instance also may select each pixel in left and right of current pixel point in the judgement of above-mentioned boundary point, or each three pixels in left and right participate in calculating and judging.Or adopt other border search method such as inner boundary searching algorithm etc., broadly fall into protection scope of the present invention.
S300: set the target width horizontal coordinate spacing as adjacent two boundary points of bar-shaped zone；
Specifically, this step chooses the horizontal coordinate spacing of adjacent two boundary points as target width, and is preserved by target width.The chosen area of target width can be beginning a line of a frame testing image or a few row, it is possible to for other optional positions.Target width is tried to achieve by the relative coordinate of adjacent two boundary point horizontal directions, such as select the horizontal coordinate spacing of the first boundary point of Nth row and the second boundary adjacent with the first boundary point point as target width, if the horizontal position coordinate of the first boundary point is i, the horizontal position coordinate of the second boundary point is j, then target width is | j-i |.
S400: calculating the horizontal coordinate spacing of adjacent two boundary points and the difference of described target width line by line one by one, whether the described difference of statistics each row is in preset range, and if so, then this row meets bar paten feature；If the line number exceeding preset ratio meets bar paten feature, then described tested pattern is bar paten；Otherwise, tested pattern is non-bar paten.
Specifically, the horizontal coordinate distance computation method of adjacent two boundary points of this step is referred to the computational methods of step S300.Owing to the judgement of each row is all that timesharing carries out, so for the accuracy improving a frame pattern detection, the present embodiment is after choosing suitable target width, the horizontal coordinate spacing of every pair of adjacent boundary point and target width are contrasted by each provisional capital, judge that whether both differences are zero or less than preset value, if it is zero or less than preset value, then the difference of this row is in preset range that this row has more than a certain proportion of difference, it was shown that this journey meets bar paten feature.Wherein said ratio could be arranged to more than 80%, naturally it is also possible to is set to other suitable percentage ratio.For example, it is assumed that tested pattern is K rank colour bar patterns, as long as then having equal more than the result on 0.8K rank or less than preset value, then it is assumed that current line meets bar paten feature.Wherein, described preset value can rule of thumb be arranged, and generally may be configured as 1 to 3 coordinate points, naturally it is also possible to is set to other suitable numerical value.
This step carries out contrast line by line one by one boundary point and judges, often contrast a line, namely judge whether this journey meets bar paten feature, until all row judge complete, finally statistics has how many row to meet bar paten feature, if exceeding preset ratio, then judge that this tested pattern is bar paten, be otherwise non-bar paten.Wherein, described preset ratio could be arranged to more than 90%, it is of course also possible to be set to other suitable percentage ratio.
The present embodiment is by converting yuv format to by unified for tested pattern, obtain luminance component Y, boundary point is searched for according to luminance component Y, target width is determined by calculating the horizontal coordinate spacing of adjacent two boundary points, the selection mode making target width is based on tested pattern and selects, this kind of mode is more flexible, tested pattern is made to be no longer limited to the bar paten of fixing exponent number, the bar paten of any exponent number can also be detected, and all scanning provisional capitals of a two field picture are detected, improves the accuracy judging bar paten.
As in figure 2 it is shown, be the bar paten detection method block diagram of another embodiment of the present invention, the detection method of the present embodiment, the basis of above-described embodiment further comprises the steps of:
S500: judge that in tested pattern, whether the chromatic value of each pixel is less than default Chroma threshold, be if so, then the pixel without colourity；If the pixel accounting without colourity exceedes default threshold value, then tested pattern is achromatic color pattern, is otherwise multicolour pattern.
After step S400 judges that tested pattern is bar paten, also include step S600: judge that described achromatic color pattern is the grey scale pattern without colourity, it is judged that described multicolour pattern is colour bar pattern.
Specifically, after tested pattern is converted to yuv format, obtaining the chromatic value C of each pixel, this step judges that whether the chromatic value C of each pixel in tested pattern is less than default threshold value one by one.The artwork master of general standard or GTG without in chromatic graph, during netrual colour, chromatic value C is 0, but in practice owing to the video signal such as CVBS signal etc. of input is all containing noise dither after the transmission system of series of complex, so needing the Rule of judgment arranging a threshold value more than 0 as netrual colour, this threshold value generally dynamically can be configured by systems soft ware.If pixel chromatic value C is less than the threshold value set, then this pixel is non-chroma point, and in the statistics one tested pattern of frame, whether non-chroma point proportion is more than default ratio, if greater than this ratio preset, then tested pattern is the achromatic color pattern without chrominance information, is otherwise multicolour pattern.Judgement according to step 400 again, if meeting both had been bar paten simultaneously, is again multicolour pattern, then tested pattern is colour bar pattern；If bar paten and be achromatic color pattern, then be the grey scale pattern without colourity.If judging in step 400, tested pattern is not bar paten, then no longer judge that it is the grey scale pattern without colourity or colour bar pattern.Wherein, described default ratio could be arranged to 95%, or other suitable ratios.
It addition, in the present embodiment, in order to strengthen the reliability of the target width chosen in step S300, target width is set to the intermediate value of the horizontal coordinate spacing of the multipair adjacent boundary point of multirow.Such as, it is K1 by the horizontal coordinate spacing of the 1st of the 1st row the pair of adjacent boundary point, the horizontal coordinate spacing of the 1st the 2nd pair of adjacent boundary point of row is K2, the horizontal coordinate spacing of the 3rd pair of adjacent boundary point of the 3rd row is K3, the horizontal coordinate spacing of the 4th pair of adjacent boundary point of the 4th row is K4, and target width is K1, K2, the intermediate value of K3, K4.Specific implementation, can adopt median filter, and the horizontal coordinate spacing of the adjacent boundary of each row point is carried out medium filtering and preserved.
Furthermore it is also possible to target width to be set to the meansigma methods of the horizontal coordinate spacing of the multipair adjacent boundary point of multirow.Such as, it is K1 by the horizontal coordinate spacing of the 1st of the 1st row the pair of adjacent boundary point, the horizontal coordinate spacing of the 1st the 2nd pair of adjacent boundary point of row is K2, the horizontal coordinate spacing of the 3rd pair of adjacent boundary point of the 3rd row is K3, the horizontal coordinate spacing of the 4th pair of adjacent boundary point of the 4th row is K4, and target width is K1, K2, the meansigma methods of K3, K4.Specific implementation, can adopt mean filter, and the horizontal coordinate spacing of the adjacent boundary of each row point is carried out mean filter and preserved.
The present embodiment judges on the basis that tested pattern is bar paten at above-described embodiment, also by whether the chromatic value contrasting each pixel exceedes certain Chroma threshold to judge that tested pattern is colour bar pattern or the grey scale pattern without colourity, so, have great convenience for the user the understanding to system equipment situation, and input video source is carried out color correction and gray level white balance correction, so that Computer Vision obtains better color representation ability, better effects if, making the corresponding product more market competitiveness, engineer applied is worth greatly.
Again as it is shown on figure 3, be the bar paten structure of the detecting device figure of one embodiment of the invention, the detecting device of the present embodiment includes format converting module, search module, target width determine module, width calculation module and judge module；
It is yuv format that described format converting module is used for the image format conversion of tested pattern.
Specifically, because the tested pattern source form of input is different, it is necessary to be converted into yuv format by unified for the video image of various forms, to obtain the brightness value Y and chromatic value C of each pixel of tested pattern.The brightness value Y of each pixel is exported to search module by format converting module.Wherein yuv format includes YUV420, YUV422 etc..
Described search module, for the tested pattern after horizontal sweep conversion, searches for boundary point.
Specifically, search module can realize boundary point search by boundary search algorithm, such as luminance detail value searching algorithm, the detail section in a tested pattern of frame is judged by application searches algorithm on luminance component Y, and namely the obvious region of details is regarded as the region that border exists.Such as, the region having vertical bar is detail areas, and solid color regions is smooth area.Tested pattern after search module horizontal sweep conversion, judge whether each pixel is boundary point, by detecting the brightness value sum of two pixels in the left side of current pixel point, difference with the brightness value sum of two pixels in the right, judge whether this difference judges more than default luminance threshold, if so, then this pixel is boundary point, and otherwise this pixel is large area smooth region.Wherein luminance threshold dynamically can be configured by systems soft ware.
The specific embodiment of the invention is not limited to above-mentioned border search method, for instance in the judgement of above-mentioned boundary point, each pixel in left and right of search module also optional current pixel point, or each three pixels in left and right participate in calculating and judging.Or adopt other border search method such as inner boundary searching algorithm etc., broadly fall into protection scope of the present invention.
Described target width determines that module for being set to the target width of bar-shaped zone by the horizontal coordinate spacing of adjacent two boundary points.
Specifically, the horizontal coordinate spacing of arbitrary neighborhood two boundary point can be chosen as target width, and target width is preserved.The chosen area of target width can be beginning a line of a frame pattern to be measured or a few row, it is possible to for other optional positions.Target width is tried to achieve by the relative coordinate of adjacent two boundary point horizontal directions, such as select the horizontal coordinate spacing of the first boundary point of Nth row and the second boundary adjacent with the first boundary point point as target width, if the horizontal position coordinate of the first boundary point is i, the horizontal position coordinate of the second boundary point is j, then target width is | j-i |.
Described width calculation module is for calculating the horizontal coordinate spacing of adjacent two boundary points and the difference of described target width line by line one by one, and whether the described difference of statistics each row is in preset range, and if so, then this row meets bar paten feature.
Specifically, width calculation module calculates the horizontal coordinate spacing of adjacent two boundary points and the difference of described target width line by line one by one, width calculation module calculates the method for horizontal coordinate spacing of adjacent two boundary points and determines that module is identical with above-mentioned target width, and two modules can share the calculating logic of the horizontal coordinate spacing of adjacent two boundary points.After target width determines that module chooses suitable target width, the horizontal coordinate spacing of every pair of adjacent boundary point of every a line and target width are contrasted by width calculation module line by line, judge that whether both differences are zero or less than preset value, if it is zero or less than preset value that this row has more than a certain proportion of difference, then the difference of this row is in preset range, it was shown that this journey meets bar paten feature.
In the present embodiment, width calculation module can include computing module and comparison module, and described computing module is for calculating the horizontal coordinate spacing of adjacent two boundary points and the difference of described target width line by line one by one；Described comparison module is for judging that whether difference that described computing module draws is zero or less than preset value, if it is zero or less than preset value that current line exceedes a certain proportion of difference, then the difference of current line is in preset range, and otherwise, the difference of current line is not in preset range.
With target width, described computing module can determine that module shares width calculation logic.After computing module has calculated a difference, described comparison module judges that whether this difference is zero or less than preset value, if it is zero or less than preset value that current line exceedes a certain proportion of difference, then the difference of current line is in preset range, this row meets bar paten feature, otherwise, the difference of current line, not in preset range, does not meet bar paten feature.Wherein said ratio could be arranged to more than 80%, naturally it is also possible to is set to other suitable percentage ratio.For example, it is assumed that tested pattern is K rank colour bar patterns, as long as then having equal more than the result on 0.8K rank or less than preset value, then it is assumed that current line meets bar paten feature.Wherein, described preset value can rule of thumb be arranged, and generally may be configured as 1 to 3 coordinate points or other suitable numerical value.Often contrast a line, namely judged whether this journey meets bar paten feature, until all row judge complete.
Described judge module for judging whether that the line number exceeding preset ratio meets bar paten feature, be if so, then described tested pattern is bar paten；Otherwise, tested pattern is non-bar paten.
Specifically, after width calculation module has judged whether all row meet bar paten feature, it is judged that module has finally added up how many row meets bar paten feature, if the line number exceeding preset ratio meets bar paten feature, then judge that this tested pattern is bar paten, be otherwise non-bar paten.Wherein, described preset ratio could be arranged to more than 90%, or other suitable percentage ratios.
The present embodiment is by converting yuv format to by unified for tested pattern, obtain luminance component Y, boundary point is searched for according to luminance component Y, target width is determined by calculating the horizontal coordinate spacing of adjacent two boundary points, the selection mode making target width is based on tested pattern and selects, this kind of mode is more flexible, tested pattern is made to be no longer limited to the bar paten of fixing exponent number, the bar paten of any exponent number can also be detected, and all scanning provisional capitals of a two field picture are detected, improves the accuracy judging bar paten.
Again as shown in Figure 4, for the bar paten structure of the detecting device figure of another embodiment of the present invention, the detecting device of the present embodiment also includes chrominance distortion module and colourity judge module on the basis of above-described embodiment.Described chrominance distortion module obtains the chromatic value of each pixel in the tested pattern that described format converting module obtains, judge that in tested pattern, whether the chromatic value of each pixel is less than default Chroma threshold, if so, it is then the pixel without colourity, and adds up the pixel proportion without colourity；Described colourity judge module is for judging whether the described pixel proportion without colourity exceedes default threshold value, and if so, then tested pattern is the achromatic color pattern without carrier chrominance signal, is otherwise multicolour pattern.
Described judge module is additionally operable to after judging that tested pattern is bar paten, it is judged that described achromatic color pattern is the grey scale pattern without colourity, it is judged that described multicolour pattern is colour bar pattern.
Specifically, after described format converting module changes the form of tested pattern, obtaining the chromatic value C of each pixel, described chrominance distortion module is arrived in output.Chrominance distortion module judges that whether the chromatic value C of each pixel in tested pattern is less than default threshold value one by one.The artwork master of general standard or GTG without in chromatic graph, during netrual colour, chromatic value C is 0, but in practice owing to the video signal such as CVBS signal etc. of input is all containing noise dither after the transmission system of series of complex, so needing the Rule of judgment arranging a threshold value more than 0 as netrual colour, this threshold value generally dynamically can be configured by systems soft ware.If pixel chromatic value C is less than the threshold value set, then this pixel is non-chroma point, and non-chroma point proportion in the statistics one tested pattern of frame.Colourity judge module judges that whether this ratio is more than default ratio, if greater than default ratio, then judges that tested pattern is the achromatic color pattern without colourity, is otherwise multicolour pattern.Wherein, described default ratio could be arranged to 95%, or other suitable ratios.
It addition, in the present embodiment, in order to strengthen the reliability of the target width chosen, described target width determines that target width is set to the intermediate value of the horizontal coordinate spacing of the multipair adjacent boundary point of multirow by module.Such as, it is K1 by the horizontal coordinate spacing of the 1st of the 1st row the pair of adjacent boundary point, the horizontal coordinate spacing of the 1st the 2nd pair of adjacent boundary point of row is K2, the horizontal coordinate spacing of the 3rd pair of adjacent boundary point of the 3rd row is K3, the horizontal coordinate spacing of the 4th pair of adjacent boundary point of the 4th row is K4, and target width is K1, K2, the intermediate value of K3, K4.Specific implementation, can adopt median filter, the horizontal coordinate spacing of the adjacent boundary of each row point carries out medium filtering and preserves as target width.
It addition, described target width determines that target width can also be set to the meansigma methods of the horizontal coordinate spacing of the multipair adjacent boundary point of multirow by module.Such as, it is K1 by the horizontal coordinate spacing of the 1st of the 1st row the pair of adjacent boundary point, the horizontal coordinate spacing of the 1st the 2nd pair of adjacent boundary point of row is K2, the horizontal coordinate spacing of the 3rd pair of adjacent boundary point of the 3rd row is K3, the horizontal coordinate spacing of the 4th pair of adjacent boundary point of the 4th row is K4, and target width is K1, K2, the meansigma methods of K3, K4.Specific implementation, can adopt mean filter, and the horizontal coordinate spacing of the adjacent boundary of each row point is carried out mean filter and preserved.
The present embodiment judges on the basis that tested pattern is bar paten at above-described embodiment, also by whether the chromatic value contrasting each pixel exceedes certain Chroma threshold to judge that tested pattern is colour bar pattern or the grey scale pattern without colourity, so, have great convenience for the user the understanding to system equipment situation, and input video source is carried out color correction and gray level white balance correction, so that Computer Vision obtains better color representation ability, better effects if, making the corresponding product more market competitiveness, engineer applied is worth greatly.
More than lift preferred embodiment; the object, technical solutions and advantages of the present invention have been further described; it is it should be understood that; the foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all within the spirit and principles in the present invention; any amendment of being made, equivalent replacement, improvement etc.; should be included within protection scope of the present invention, the interest field that the present invention advocates should be as the criterion described in patent application scope, but not is only limitted to above-described embodiment.
Claims (10)
1. a bar paten detection method, it is characterised in that comprise the following steps:
It is yuv format by the image format conversion of tested pattern；
Tested pattern after horizontal sweep conversion, searches for boundary point；
Set the target width horizontal coordinate spacing as adjacent two boundary points of bar-shaped zone；
Calculating the horizontal coordinate spacing of adjacent two boundary points and the difference of described target width line by line one by one, whether the described difference of statistics each row is in preset range, and if so, then this row meets bar paten feature；If the line number exceeding preset ratio meets bar paten feature, then described tested pattern is bar paten；Otherwise, tested pattern is non-bar paten.
2. bar paten detection method as claimed in claim 1, it is characterised in that also comprise the steps:
Judge that in tested pattern, whether the chromatic value of each pixel is less than default Chroma threshold, be if so, then the pixel without colourity；If the pixel accounting without colourity exceedes default threshold value, then tested pattern is achromatic color pattern, is otherwise multicolour pattern；
After the tested pattern of described judgement is bar paten, then can determine whether that described achromatic color pattern is the grey scale pattern without colourity, described multicolour pattern is colour bar pattern.
3. bar paten detection method as claimed in claim 1, it is characterised in that the described difference of the described statistics each row whether step in default scope includes:
Judge that whether the difference of horizontal coordinate spacing and the target width of every pair of adjacent boundary point of current line is zero or less than preset value, if it is zero or less than preset value that current line exceedes a certain proportion of difference, then the difference of current line is in preset range, and otherwise, the difference of current line is not in preset range.
4. bar paten detection method as claimed in claim 1, it is characterised in that the step of described search boundary point includes utilizing luminance detail value searching algorithm or inner boundary searching algorithm search boundary point.
5. bar paten detection method as claimed in claim 1, it is characterised in that the described target width setting bar-shaped zone includes as the step of the horizontal coordinate spacing of adjacent two boundary points:
Calculate the horizontal coordinate spacing of the multipair adjacent boundary point of more than two row, take intermediate value or meansigma methods is target width.
6. a bar paten detecting device, it is characterised in that include format converting module, search module, target width determine module, width calculation module and judge module；
It is yuv format that described format converting module is used for the image format conversion of tested pattern；
Described search module, for the tested pattern after horizontal sweep conversion, searches for boundary point；
Described target width determines that module for being set to the target width of bar-shaped zone by the horizontal coordinate spacing of adjacent two boundary points；
Described width calculation module is for calculating the horizontal coordinate spacing of adjacent two boundary points and the difference of described target width line by line one by one, and whether the described difference of statistics each row is in preset range, and if so, then this row meets bar paten feature；
Described judge module for judging whether that the line number exceeding preset ratio meets bar paten feature, be if so, then described tested pattern is bar paten；Otherwise, tested pattern is non-bar paten.
7. bar paten detecting device as claimed in claim 6, it is characterized in that, also include chrominance distortion module and colourity judge module, described chrominance distortion module is for judging that in tested pattern, whether the chromatic value of each pixel is less than default Chroma threshold, if, it is then the pixel without colourity, and adds up the pixel proportion without colourity；Described colourity judge module is for judging whether the described pixel proportion without colourity exceedes default threshold value, and if so, then tested pattern is achromatic color pattern, is otherwise multicolour pattern；
Described judge module is additionally operable to after judging that tested pattern is bar paten, it is judged that described achromatic color pattern is the grey scale pattern without colourity, it is judged that described multicolour pattern is colour bar pattern.
8. bar paten detecting device as claimed in claim 6, it is characterized in that, described width calculation module includes computing module and comparison module, and described computing module is for calculating the horizontal coordinate spacing of adjacent two boundary points and the difference of described target width line by line one by one；
Described comparison module is for judging that whether difference that described computing module draws is zero or less than preset value, if it is zero or less than preset value that current line exceedes a certain proportion of difference, then the difference of current line is in preset range, and otherwise, the difference of current line is not in preset range.
9. bar paten detecting device as claimed in claim 6, it is characterised in that described search module adopts luminance detail value searching algorithm or inner boundary searching algorithm search boundary point.
10. bar paten detecting device as claimed in claim 6, it is characterised in that described target width determines that module passes through to calculate the horizontal coordinate spacing of the multipair adjacent boundary point of more than two row, and intermediate value or meansigma methods are set to target width.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410745508.4A CN105744260B (en) | 2014-12-09 | 2014-12-09 | A kind of bar paten detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410745508.4A CN105744260B (en) | 2014-12-09 | 2014-12-09 | A kind of bar paten detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105744260A true CN105744260A (en) | 2016-07-06 |
| CN105744260B CN105744260B (en) | 2018-10-12 |
Family
ID=56237901
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410745508.4A Active CN105744260B (en) | 2014-12-09 | 2014-12-09 | A kind of bar paten detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105744260B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543624A (en) * | 2018-05-28 | 2019-12-06 | 珠海金山办公软件有限公司 | identification method and device for check boxes in PDF document and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101742360A (en) * | 2008-11-11 | 2010-06-16 | 康佳集团股份有限公司 | Display equipment capable of identifying standard color bar signals and identification method used by same |
| CN101815222A (en) * | 2009-02-25 | 2010-08-25 | 北大方正集团有限公司 | Video color bar detecting method and device |
| CN101814188A (en) * | 2009-06-16 | 2010-08-25 | 黑龙江工程学院 | Color fringe edge sub-pixel detection method based on color component intersection point |
| CN102006499A (en) * | 2010-12-10 | 2011-04-06 | 北京中科大洋科技发展股份有限公司 | Method for detecting video and audio quality of digital television program file |
| JP2013225795A (en) * | 2012-04-23 | 2013-10-31 | Kddi Corp | Abnormal content detection system, method and computer program |
-
2014
- 2014-12-09 CN CN201410745508.4A patent/CN105744260B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101742360A (en) * | 2008-11-11 | 2010-06-16 | 康佳集团股份有限公司 | Display equipment capable of identifying standard color bar signals and identification method used by same |
| CN101815222A (en) * | 2009-02-25 | 2010-08-25 | 北大方正集团有限公司 | Video color bar detecting method and device |
| CN101814188A (en) * | 2009-06-16 | 2010-08-25 | 黑龙江工程学院 | Color fringe edge sub-pixel detection method based on color component intersection point |
| CN102006499A (en) * | 2010-12-10 | 2011-04-06 | 北京中科大洋科技发展股份有限公司 | Method for detecting video and audio quality of digital television program file |
| JP2013225795A (en) * | 2012-04-23 | 2013-10-31 | Kddi Corp | Abnormal content detection system, method and computer program |
Non-Patent Citations (1)
| Title |
|---|
| 伍以文: "音视频节目技术审片***的研究与实现", 《中国优秀硕士学位论文（信息科技辑）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543624A (en) * | 2018-05-28 | 2019-12-06 | 珠海金山办公软件有限公司 | identification method and device for check boxes in PDF document and electronic equipment |
| CN110543624B (en) * | 2018-05-28 | 2023-06-23 | 珠海金山办公软件有限公司 | Method and device for identifying check boxes in PDF document and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105744260B (en) | 2018-10-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3534326A1 (en) | Method and apparatus for merging infrared image and visible light image | |
| CN102273208B (en) | Image processing device and image processing method | |
| CN105678700B (en) | Image interpolation method and system based on prediction gradient | |
| CN101080022B (en) | Imaging device | |
| US9478192B2 (en) | Image processing method | |
| CN101770646B (en) | Edge detection method based on Bayer RGB images | |
| CN102263985B (en) | Quality evaluation method, device and system of stereographic projection device | |
| CN103067734B (en) | Video image color cast detecting method of video quality diagnostic system | |
| US20080181494A1 (en) | Image Processing Method, Display Image Processing Method, Image Processing Device, Image Processing Program, and Integrated Circuit Containing the Image Processing Device | |
| CN100571335C (en) | Image syncretizing effect real-time estimating method and device based on pixel space relativity | |
| EP3358844A1 (en) | Method and apparatus for processing an image property map | |
| CN102819850A (en) | Method for detecting edge of color image on basis of local self-adaption color difference threshold | |
| JP2015106890A (en) | Pixel interpolation device, imaging apparatus, program and integrated circuit | |
| CN104618703A (en) | White balance adjusting method | |
| CN105430357B (en) | The demosaicing methods and device of imaging sensor | |
| CN101923455B (en) | Method for displaying and analyzing digital image in YUV format | |
| TWI544785B (en) | Image downsampling apparatus and method | |
| KR20120015980A (en) | Method, system and computer program product for object color correction | |
| US20110279715A1 (en) | Blemish detection sytem and method | |
| TWI758630B (en) | Image zooming method and device | |
| CN105744260A (en) | Bar pattern detection method and device | |
| TWI404408B (en) | Image processing apparatus and image processing method | |
| CN104967840B (en) | A kind of exceeded detection method and device of video image | |
| CN103763498B (en) | HDMI signal processing systems and processing method | |
| CN103002230B (en) | Adaptability pixel compensation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| CB02 | Change of applicant information |
Address after: 518057 4 room 406-421, phase 1, two software park, Nanshan District science and technology, Guangdong, Shenzhen Applicant after: Shenzhen Shenyang electronic Limited by Share Ltd Address before: 518057 4 room 406-421, phase 1, two software park, Nanshan District science and technology, Guangdong, Shenzhen Applicant before: Aike Chuangxin Microelectronic Co., Ltd. |
|
| CB02 | Change of applicant information | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |