CN102129693A - Image vision significance calculation method based on color histogram and global contrast - Google Patents
Image vision significance calculation method based on color histogram and global contrast Download PDFInfo
- Publication number
- CN102129693A CN102129693A CN2011100625201A CN201110062520A CN102129693A CN 102129693 A CN102129693 A CN 102129693A CN 2011100625201 A CN2011100625201 A CN 2011100625201A CN 201110062520 A CN201110062520 A CN 201110062520A CN 102129693 A CN102129693 A CN 102129693A
- Authority
- CN
- China
- Prior art keywords
- color
- representative
- representative color
- conspicuousness
- frequency
- 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
- 230000004438 eyesight Effects 0.000 title claims abstract description 26
- 238000004364 calculation method Methods 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 35
- 239000003086 colorant Substances 0.000 claims abstract description 12
- 230000000717 retained effect Effects 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 description 9
- 238000001514 detection method Methods 0.000 description 6
- 238000002474 experimental method Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 239000004744 fabric Substances 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
The invention discloses an image vision significance calculation method based on a color histogram and global contrast, comprising the steps of: S1, quantizing color space to obtain a group of representative colors; S2, calculating the occurrence frequency of colors corresponding to the representative colors in an input image to constitute a histogram; S3, calculating the significance value of the representative colors according to the difference between each representative color and other representative colors; and S4, for each representative color, giving the significance value of the representative colors corresponding pixels. By means of the method provided by the invention, the significance value of image pixels can be rapidly and effectively analyzed.
Description
Technical field
The present invention relates to technical field of image processing, particularly a kind of image vision conspicuousness computing method based on color histogram and global contrast.
Background technology
Vision attention is to help the human visual system to discern a kind of important mechanisms of scene accurately and effectively.The salient region that obtains in the image by computing method is important research project of computer vision field.It can help image processing system rational distributes calculation resources in subsequent processing steps.Conspicuousness figure (Saliency map) is widely used in the middle of many computer visions application, cut apart the image zoom (patent 200910092756), image retrieval (patent 200910081069) of the compression of (patent 200910046276,200910081069), object identification, adapting to image, content erotic etc. as attention object.
It is a problem that enjoys the researcher to pay close attention to that the image vision conspicuousness detects all the time.Theoretical research about vision attention is divided into two classes with vision attention: fast, task conspicuousness irrelevant, data-driven detects with slow, task is relevant, the detection of the conspicuousness of target drives.Method involved in the present invention belongs to last class.Physiologic Studies shows, the stimulation that has higher contrast in the human vision cell preferential answering perception field.The vision significance that available data drives detects research and comes the computation vision conspicuousness by the contrast of calculating various forms of picture materials and scene mostly.Introduce for convenience, further such research is subdivided into two subclasses: based on the method for local contrast with based on the method for global contrast.
Method based on local contrast is calculated conspicuousness than the rare degree of small neighbourhood by image-region relatively at it.People such as Itti have proposed " A model of saliency-based visual attention for rapid scene analysis " in 1998.This method defines the image conspicuousness by center between the multi-scale image feature and neighborhood difference.Ma and Zhang and proposed " Contrast-based image attention analysis by using fuzzy growing " in 2003.This method obtains conspicuousness figure by the local contrast analysis.People such as Liu have proposed " Learning to detect a salient object " in 2007.This method finds color space distribution, multiple dimensioned contrast, conspicuousness detection method results' such as center neighborhood histogram difference optimum combination weights by mode of learning.People such as Goferman carry out modeling to bottom clue, overall consideration, organization regulation and high-level characteristic in its work in 2010 " Context-aware saliency detection ".The result of these partial approaches produces higher conspicuousness value usually near object edge, rather than outstanding uniformly whole vision significance object.
Method based on global contrast is estimated its conspicuousness by the difference of tolerance image-region and entire image.Zhai and Shah have proposed " Visual attention detection in video sequences using spatiotemporal cues " in 2006.This method is calculated the conspicuousness value of this pixel by the luminance difference of a pixel and other all images pixel.Based on the consideration of efficient aspect, this method is only utilized the monochrome information of image, has ignored the differentiation in other Color Channel.People such as Achanta have proposed " Frequency-tuned salient region detection " in 2009.This method is obtained the conspicuousness value by the difference of calculating between each pixel and the average color.Yet this simple method is not enough to the changeable natural image of effective Analysis of Complex.
This field domestic relevant patent at present has: based on method for automatically detecting obvious object sequence (patent No. 200810150324) in the video of study.This method is handled a width of cloth picture needs some seconds time usually, is difficult to satisfy a lot of true application demands of handling.
Summary of the invention
(1) technical matters that will solve
The technical problem to be solved in the present invention is: how fast and effectively the conspicuousness value of analysis image pixel makes that the important objects zone can be highlighted uniformly in the image.
(2) technical scheme
For solving the problems of the technologies described above, the invention provides a kind of image vision conspicuousness computing method based on color histogram and global contrast, may further comprise the steps:
S1: color space is quantized to obtain one group of representative color;
S2: calculate the frequency of occurrences of color in input picture of described representative color correspondence, form a histogram;
S3: the conspicuousness value of calculating representative color according to each representative color and other representative difference in color;
S4:, give corresponding pixel with its conspicuousness value for each representative color.
Wherein, among the described step S2, from big to small representative color is sorted according to the frequency of occurrences of described representative color in input picture, the representative color of predetermined number is retained before coming, and the frequency of occurrences of all the other representative colors is added on the frequency of occurrences of the most close representative color that is retained of color.
Wherein, the computing formula of the conspicuousness value S of representative color is among the described step S3:
Wherein, C is the representative color set that is retained among the step S2, and c is arbitrary representative color in this set, c
iBe other the representative color except that c, f
iBe c
iThe frequency of occurrences, D (c, c
i) be c, c
iEuclidean distance in color space.
Wherein, comprise further also between step S3 and the S4 that following process improves testing result: in color space the conspicuousness value of described representative color is carried out smoothly, concrete smoothing formula is:
Wherein, S ' is the conspicuousness value of the representative color c after level and smooth (c), C
mBe the set that the m the most close with the color value of representative color c representative color formed, w
iBe weights, c
iClose more then w with c
iBig more.
(3) beneficial effect
The method utilization of the vision significance value that is used for the computed image pixel proposed by the invention utilizes the difference between each pixel and the rest of pixels to calculate its conspicuousness value; For speed-up computation, this method is by color quantizing and select more frequent color to select one group of representative color; This method can further be improved the conspicuousness testing result by a kind of color space smoothing method, fast and effeciently the conspicuousness value of analysis image pixel.Obtained the result who obviously is better than than classic method on the existing in the world maximum test set of this method.
Description of drawings
Fig. 1 is a kind of image vision conspicuousness computing method process flow diagram based on color histogram and global contrast of the embodiment of the invention;
Fig. 2 adopts the method flow of Fig. 1 to handle the exemplary plot of a width of cloth input picture;
Representative color is chosen and the exemplary plot of corresponding occurrence frequency in the method for Fig. 3 Fig. 1;
Color space smoothly improves the exemplary plot of influence in the method for Fig. 4 Fig. 1 to testing result.
Embodiment
Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples are used to illustrate the present invention, but are not used for limiting the scope of the invention.
As shown in Figure 1, the processing flow chart for an embodiment of the image vision conspicuousness computing method that the present invention is based on color histogram and global contrast comprises:
Step S101 quantizes to obtain one group of representative color to color space.The color quantizing mode that adopts in the experiment evenly is divided into n with color space
3Part, the mean value of color is as this regional representative colors in each zone.In actual computation, the value of n is big more in the computation process, and the calculated amount of whole algorithm is just big more.In experiment, preferred n=12 just can obtain good result.
Step S102 calculates the frequency of occurrences of color in input picture of representative color correspondence, forms a histogram, and certain representative color respective pixel accounts for the frequency of occurrences of the ratio of all number of pixels in the input picture for this representativeness color.Each representative color all has a frequency.This frequency of organizing representative color appearance is called as histogram, as shown in Figure 2.In order to save computational resource, keep the more representative color of the frequency of occurrences usually, the frequency of occurrences of all the other representative colors is accumulated on the frequency of occurrences of the most close representative color that is retained of color.When selecting the more representative color of the frequency of occurrences, the frequency of occurrences of representative color is sorted from big to small.Select to be enough to cover the representative colors of certain proportion image pixel then from front to back.This ratio is chosen as 95% usually in the experiment, and the statistics of the number of the representative color of this ratio correspondence is 85 on a maximum in the world at present public data collection.
Step S103 calculates the conspicuousness value S of representative color according to each representative color and other representative difference in color, and concrete computing formula is as follows:
Wherein, C is the representative color set that is retained among the step S102, and c is arbitrary representative color in this set, c
iBe other the representative color except that c, f
iBe c
iThe frequency of occurrences, D (c, c
i) be c, c
iEuclidean distance in color space.
Step S104 for each representative color, gives corresponding pixel with its conspicuousness value.
Exemplary plot when being illustrated in figure 2 as said process and handling a width of cloth picture has comprised the frequency of occurrences and the conspicuousness histogram of representative color.
As shown in Figure 3, choose for representative color in the image vision conspicuousness computing method that the present invention is based on color histogram and global contrast and the exemplary plot of corresponding occurrence frequency.Among the figure, replace image (right figure) behind the original image (left figure) with representative color and kept and carry out the required enough information of conspicuousness detection, the while has been simplified calculated amount greatly.
As shown in Figure 4, smoothly testing result is improved the exemplary plot of influence for color space in the image vision conspicuousness computing method that the present invention is based on color histogram and global contrast.By at color space the conspicuousness value of representative color being carried out smoothly can effectively improving the testing result of vision significance.In step S103, calculate representative color S (c) afterwards, the sets definition that the representative color of the m that certain representative color c is the most close is formed is C
mThe smoothing process of this color space can be defined as:
Wherein, S ' is the conspicuousness value of the representative color c after level and smooth (c), w
iBe weights, c
iClose more then w with c
iBig more.Before and after left figure and right figure have provided respectively and have improved among Fig. 4, an example of the variation of the conspicuousness value of representative colors and the conspicuousness figure of input picture.In experiment the value of m be generally representative color number 25% just can obtain reasonable result.
Image conspicuousness computing method disclosed by the invention define its conspicuousness value by certain pixel and all the other all color of pixel differences.For fear of too huge calculation cost, this method is chosen representative color by color quantizing and the analysis of color occurrence frequency and is come acceleration problem to find the solution.Randomness in the color quantizing process, this method can be by the level and smooth machine-processed controlled quentity controlled variable error of a kind of color space.Obtained the result who obviously is better than than classic method on the existing in the world maximum test set of this method.The present invention is the vision significance zone in the analysis image automatically, applications such as analysis result can be applied to that important goal is cut apart, the image zoom of object identification, adapting to image compression, content erotic and image retrieval.
Above embodiment only is used to illustrate the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; under the situation that does not break away from the spirit and scope of the present invention; can also make various variations and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.
Claims (4)
1. the image vision conspicuousness computing method based on color histogram and global contrast is characterized in that, may further comprise the steps:
S1: color space is quantized to obtain one group of representative color;
S2: calculate the frequency of occurrences of color in input picture of described representative color correspondence, form a histogram;
S3: the conspicuousness value of calculating representative color according to each representative color and other representative difference in color;
S4:, give corresponding pixel with its conspicuousness value for each representative color.
2. the image vision conspicuousness computing method based on color histogram and global contrast as claimed in claim 1, it is characterized in that, among the described step S2, from big to small representative color is sorted according to the frequency of occurrences of described representative color in input picture, the representative color of predetermined number is retained before coming, and the frequency of occurrences of all the other representative colors is added on the frequency of occurrences of the most close representative color that is retained of color.
3. the image vision conspicuousness computing method based on color histogram and global contrast as claimed in claim 1 is characterized in that, the computing formula of the conspicuousness value S of representative color is among the described step S3:
Wherein, C is the representative color set that is retained among the step S2, and c is arbitrary representative color in this set, c
iBe other the representative color except that c, f
iBe c
iThe frequency of occurrences, D (c, c
i) be c, c
iEuclidean distance in color space.
4. the image vision conspicuousness computing method based on color histogram and global contrast as claimed in claim 3, it is characterized in that, comprise further also between step S3 and the S4 that following process improves testing result: in color space the conspicuousness value of described representative color is carried out smoothly, concrete smoothing formula is:
Wherein, S ' is the conspicuousness value of the representative color c after level and smooth (c), C
mBe the set that the m the most close with the color value of representative color c representative color formed, w
iBe weights, c
iClose more then w with c
iBig more.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100625201A CN102129693B (en) | 2011-03-15 | 2011-03-15 | Image vision significance calculation method based on color histogram and global contrast |
| PCT/CN2011/000690 WO2012122682A1 (en) | 2011-03-15 | 2011-04-20 | Method for calculating image visual saliency based on color histogram and overall contrast |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100625201A CN102129693B (en) | 2011-03-15 | 2011-03-15 | Image vision significance calculation method based on color histogram and global contrast |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102129693A true CN102129693A (en) | 2011-07-20 |
| CN102129693B CN102129693B (en) | 2012-07-25 |
Family
ID=44267768
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011100625201A Active CN102129693B (en) | 2011-03-15 | 2011-03-15 | Image vision significance calculation method based on color histogram and global contrast |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102129693B (en) |
| WO (1) | WO2012122682A1 (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102360490A (en) * | 2011-09-30 | 2012-02-22 | 北京航空航天大学 | Color conversion and editing propagation-based method for enhancing seasonal feature of image |
| CN102637253A (en) * | 2011-12-30 | 2012-08-15 | 清华大学 | Video foreground object extracting method based on visual saliency and superpixel division |
| CN102799882A (en) * | 2012-07-09 | 2012-11-28 | 武汉市科迅智能交通设备有限公司 | License plate positioning method based on visual saliency |
| CN102938139A (en) * | 2012-11-09 | 2013-02-20 | 清华大学 | Automatic synthesis method for fault finding game images |
| CN102999926A (en) * | 2012-11-12 | 2013-03-27 | 北京交通大学 | Low-level feature integration based image vision distinctiveness computing method |
| CN103810707A (en) * | 2014-01-28 | 2014-05-21 | 华东理工大学 | Mobile visual focus based image vision salient detection method |
| WO2014169822A1 (en) * | 2013-04-19 | 2014-10-23 | 广东图图搜网络科技有限公司 | Image segmentation method and system |
| CN104951440A (en) * | 2014-03-24 | 2015-09-30 | 联想(北京)有限公司 | Image processing method and electronic device |
| CN104952084A (en) * | 2015-05-26 | 2015-09-30 | 深圳市万普拉斯科技有限公司 | Color quantization method and system |
| WO2015196715A1 (en) * | 2014-06-24 | 2015-12-30 | 小米科技有限责任公司 | Image retargeting method and device and terminal |
| CN106056579A (en) * | 2016-05-20 | 2016-10-26 | 南京邮电大学 | Saliency detection method based on background contrast |
| WO2016197705A1 (en) * | 2015-06-09 | 2016-12-15 | 中兴通讯股份有限公司 | Image processing method and device |
| US9665925B2 (en) | 2014-06-24 | 2017-05-30 | Xiaomi Inc. | Method and terminal device for retargeting images |
| CN107784662A (en) * | 2017-11-14 | 2018-03-09 | 郑州布恩科技有限公司 | A kind of image object significance measure method |
| CN108737821A (en) * | 2018-04-25 | 2018-11-02 | 中国人民解放军军事科学院军事医学研究院 | The quick pre-selection method in video interest region based on multichannel shallow-layer feature and system |
| CN109214420A (en) * | 2018-07-27 | 2019-01-15 | 北京工商大学 | The high texture image classification method and system of view-based access control model conspicuousness detection |
| CN109410171A (en) * | 2018-09-14 | 2019-03-01 | 安徽三联学院 | A kind of target conspicuousness detection method for rainy day image |
| CN109544568A (en) * | 2018-11-30 | 2019-03-29 | 长沙理工大学 | Destination image partition method, device and equipment |
| CN109711399A (en) * | 2018-11-05 | 2019-05-03 | 北京三快在线科技有限公司 | Shop recognition methods based on image, device, electronic equipment |
| CN111292287A (en) * | 2018-12-06 | 2020-06-16 | 宏碁股份有限公司 | Image normalization method and image processing device |
| CN111311697A (en) * | 2020-03-19 | 2020-06-19 | 北京搜狐新媒体信息技术有限公司 | Method for detecting color richness of picture and related device |
| CN112418223A (en) * | 2020-12-11 | 2021-02-26 | 互助土族自治县北山林场 | Wild animal image significance target detection method based on improved optimization |
| CN114187380A (en) * | 2022-02-17 | 2022-03-15 | 杭州并坚科技有限公司 | Color transfer method based on visual saliency and channel attention mechanism |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101520894A (en) * | 2009-02-18 | 2009-09-02 | 上海大学 | Method for extracting significant object based on region significance |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2349460B (en) * | 1999-04-29 | 2002-11-27 | Mitsubishi Electric Inf Tech | Method of representing colour images |
| CN100573523C (en) * | 2006-12-30 | 2009-12-23 | 中国科学院计算技术研究所 | A kind of image inquiry method based on marking area |
| CN100552716C (en) * | 2007-04-12 | 2009-10-21 | 上海交通大学 | Under the global abnormal signal environment based on the associating remarkable figure robust image registration method |
| US8005264B2 (en) * | 2008-06-09 | 2011-08-23 | Arcsoft, Inc. | Method of automatically detecting and tracking successive frames in a region of interesting by an electronic imaging device |
| JP2012521708A (en) * | 2009-03-26 | 2012-09-13 | コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ | Method and apparatus for correcting an image using a saliency map based on color frequency |
| CN101533512B (en) * | 2009-04-24 | 2012-05-09 | 西安电子科技大学 | Method for automatically extracting interesting image regions based on human visual attention system |
-
2011
- 2011-03-15 CN CN2011100625201A patent/CN102129693B/en active Active
- 2011-04-20 WO PCT/CN2011/000690 patent/WO2012122682A1/en active Application Filing
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101520894A (en) * | 2009-02-18 | 2009-09-02 | 上海大学 | Method for extracting significant object based on region significance |
Non-Patent Citations (2)
| Title |
|---|
| 《微电子学与计算机》 20100430 史变霞,张明新,乔小妮等. 基于颜色特征的图像检索方法 第158~161页 1-4 第27卷, 第4期 * |
| 《计算机工程与科学》 20070401 王剑峰,肖国强,江健民. 基于HSI色彩空间累加直方图的图像检索算法 第55~58页 1-4 第29卷, 第4期 * |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102360490A (en) * | 2011-09-30 | 2012-02-22 | 北京航空航天大学 | Color conversion and editing propagation-based method for enhancing seasonal feature of image |
| CN102637253A (en) * | 2011-12-30 | 2012-08-15 | 清华大学 | Video foreground object extracting method based on visual saliency and superpixel division |
| CN102799882A (en) * | 2012-07-09 | 2012-11-28 | 武汉市科迅智能交通设备有限公司 | License plate positioning method based on visual saliency |
| CN102938139B (en) * | 2012-11-09 | 2015-03-04 | 清华大学 | Automatic synthesis method for fault finding game images |
| CN102938139A (en) * | 2012-11-09 | 2013-02-20 | 清华大学 | Automatic synthesis method for fault finding game images |
| CN102999926A (en) * | 2012-11-12 | 2013-03-27 | 北京交通大学 | Low-level feature integration based image vision distinctiveness computing method |
| CN102999926B (en) * | 2012-11-12 | 2016-06-29 | 北京交通大学 | A kind of image vision significance computational methods merged based on low-level image feature |
| WO2014169822A1 (en) * | 2013-04-19 | 2014-10-23 | 广东图图搜网络科技有限公司 | Image segmentation method and system |
| CN103810707B (en) * | 2014-01-28 | 2016-08-17 | 华东理工大学 | A kind of image vision significance detection method based on moving-vision focus |
| CN103810707A (en) * | 2014-01-28 | 2014-05-21 | 华东理工大学 | Mobile visual focus based image vision salient detection method |
| CN104951440A (en) * | 2014-03-24 | 2015-09-30 | 联想(北京)有限公司 | Image processing method and electronic device |
| CN104951440B (en) * | 2014-03-24 | 2020-09-25 | 联想(北京)有限公司 | Image processing method and electronic equipment |
| WO2015196715A1 (en) * | 2014-06-24 | 2015-12-30 | 小米科技有限责任公司 | Image retargeting method and device and terminal |
| US9665925B2 (en) | 2014-06-24 | 2017-05-30 | Xiaomi Inc. | Method and terminal device for retargeting images |
| CN104952084A (en) * | 2015-05-26 | 2015-09-30 | 深圳市万普拉斯科技有限公司 | Color quantization method and system |
| WO2016197705A1 (en) * | 2015-06-09 | 2016-12-15 | 中兴通讯股份有限公司 | Image processing method and device |
| CN106056579A (en) * | 2016-05-20 | 2016-10-26 | 南京邮电大学 | Saliency detection method based on background contrast |
| CN107784662A (en) * | 2017-11-14 | 2018-03-09 | 郑州布恩科技有限公司 | A kind of image object significance measure method |
| CN107784662B (en) * | 2017-11-14 | 2021-06-11 | 郑州布恩科技有限公司 | Image target significance measurement method |
| CN108737821B (en) * | 2018-04-25 | 2020-09-04 | 中国人民解放军军事科学院军事医学研究院 | Video interest area quick preselection method and system based on multi-channel shallow feature |
| CN108737821A (en) * | 2018-04-25 | 2018-11-02 | 中国人民解放军军事科学院军事医学研究院 | The quick pre-selection method in video interest region based on multichannel shallow-layer feature and system |
| CN109214420A (en) * | 2018-07-27 | 2019-01-15 | 北京工商大学 | The high texture image classification method and system of view-based access control model conspicuousness detection |
| CN109410171A (en) * | 2018-09-14 | 2019-03-01 | 安徽三联学院 | A kind of target conspicuousness detection method for rainy day image |
| CN109711399A (en) * | 2018-11-05 | 2019-05-03 | 北京三快在线科技有限公司 | Shop recognition methods based on image, device, electronic equipment |
| CN109711399B (en) * | 2018-11-05 | 2021-04-27 | 北京三快在线科技有限公司 | Shop identification method and device based on image and electronic equipment |
| CN109544568A (en) * | 2018-11-30 | 2019-03-29 | 长沙理工大学 | Destination image partition method, device and equipment |
| CN111292287A (en) * | 2018-12-06 | 2020-06-16 | 宏碁股份有限公司 | Image normalization method and image processing device |
| CN111311697A (en) * | 2020-03-19 | 2020-06-19 | 北京搜狐新媒体信息技术有限公司 | Method for detecting color richness of picture and related device |
| CN111311697B (en) * | 2020-03-19 | 2023-11-17 | 北京搜狐新媒体信息技术有限公司 | Picture color richness detection method and related device |
| CN112418223A (en) * | 2020-12-11 | 2021-02-26 | 互助土族自治县北山林场 | Wild animal image significance target detection method based on improved optimization |
| CN114187380A (en) * | 2022-02-17 | 2022-03-15 | 杭州并坚科技有限公司 | Color transfer method based on visual saliency and channel attention mechanism |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102129693B (en) | 2012-07-25 |
| WO2012122682A1 (en) | 2012-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102129693B (en) | Image vision significance calculation method based on color histogram and global contrast | |
| CN102779338B (en) | Image processing method and image processing device | |
| KR102138950B1 (en) | Depth map generation from a monoscopic image based on combined depth cues | |
| CN103136766B (en) | A kind of object conspicuousness detection method based on color contrast and color distribution | |
| US8385654B2 (en) | Salience estimation for object-based visual attention model | |
| CN102025981B (en) | Method for detecting foreground in monitoring video | |
| CN110163076A (en) | A kind of image processing method and relevant apparatus | |
| CN103824284B (en) | Key frame extraction method based on visual attention model and system | |
| CN111260037B (en) | Convolution operation method and device of image data, electronic equipment and storage medium | |
| CN110827312B (en) | Learning method based on cooperative visual attention neural network | |
| CN106358029A (en) | Video image processing method and device | |
| CN106683110A (en) | User terminal and object tracking method and device thereof | |
| CN107704797B (en) | Real-time detection method, system and equipment based on pedestrians and vehicles in security video | |
| CN103985130A (en) | Image significance analysis method for complex texture images | |
| CN106570885A (en) | Background modeling method based on brightness and texture fusion threshold value | |
| CN114449362B (en) | Video cover selection method, device, equipment and storage medium | |
| CN117037127B (en) | Pallet distribution method based on luggage type | |
| CN108932703A (en) | Image processing method, picture processing unit and terminal device | |
| CN109741300B (en) | Image significance rapid detection method and device suitable for video coding | |
| CN110322479B (en) | Dual-core KCF target tracking method based on space-time significance | |
| CN108763491A (en) | image processing method, device and terminal device | |
| Lu et al. | Clustering based road detection method | |
| CN103942759A (en) | Three-dimensional noise reduction method and device based on Gaussian background model in fixed scene | |
| CN102496163B (en) | Background reconstruction method based on gray extremum | |
| Yin et al. | Headdress Detection Based on Saliency Map for Thangka Portrait Image. |
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 |