CN103971368A - Moving object foreground extraction method based on chromatic aberration - Google Patents
Moving object foreground extraction method based on chromatic aberration Download PDFInfo
- Publication number
- CN103971368A CN103971368A CN201410196311.XA CN201410196311A CN103971368A CN 103971368 A CN103971368 A CN 103971368A CN 201410196311 A CN201410196311 A CN 201410196311A CN 103971368 A CN103971368 A CN 103971368A
- Authority
- CN
- China
- Prior art keywords
- model
- background
- aberration
- tobin
- chromatic aberration
- 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
- 230000004075 alteration Effects 0.000 title claims abstract description 70
- 238000000605 extraction Methods 0.000 title claims abstract description 15
- 238000012544 monitoring process Methods 0.000 claims abstract description 10
- 239000003550 marker Substances 0.000 claims description 27
- 238000000034 method Methods 0.000 claims description 27
- 238000005286 illumination Methods 0.000 claims description 13
- 239000000284 extract Substances 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000012986 modification Methods 0.000 claims description 9
- 230000004048 modification Effects 0.000 claims description 9
- 238000012935 Averaging Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000006872 improvement Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000013461 design Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a moving object foreground extraction method based on chromatic aberration. The moving object foreground extraction method comprises the steps that areas where a moving object is located are extracted according to brightness; next, the areas are further sifted; the areas are compared with a chromatic aberration model built in advance; a part of each area is considered as foreground, wherein the difference value between the part of each area and the chromatic aberration model is larger than a threshold value; in this way, the influence of shadows and light on moving object foreground extraction is eliminated. The moving object foreground extraction method based on chromatic aberration can be applied to video monitoring, and effective moving object foreground extraction is realized.
Description
Technical field
The present invention relates to a kind of moving target foreground extracting method based on aberration, belong to technical field of video monitoring.
Background technology
Along with society and individual improve day by day to the requirement of safety, it is more and more universal that video monitoring has become, and only provide the video monitoring of image data more and more can not satisfy the demands, in therefore increasing video monitoring equipment, embedded Intelligent Measurement module.
The extraction of sport foreground is the first step of intelligent monitoring normally, is only afterwards and for prospect, carries out the processing such as signature analysis.
Because the video camera of monitoring occasion is normally monitored fixed location, can not move, for this feature, a kind of general moving target foreground extracting method is background subtracting method.First the method gathers image model as a setting, then the every two field picture and the background model that collect is subtracted each other, and obtains thus sport foreground.
Conventionally background subtracting method has three phases: background modeling and training, foreground detection and context update, this method in use has limitation, for example solar radiation is to the shade of casting after object, also can be used as foreground extraction out, although it is larger than actual moving object that the brightness of shaded side changes, shade is not our interested part.
The present invention proposes a kind of modified statistics moving target background extracting method based on aberration, the method can effectively suppress illumination and the impact of shade on background extracting.
Summary of the invention
The technical issues that need to address of the present invention are just to overcome the defect of prior art, and a kind of moving target foreground extracting method based on aberration is provided, and it can, in video monitoring, effectively extract moving target.
For addressing the above problem, the present invention adopts following technical scheme:
The invention provides a kind of moving target foreground extracting method based on aberration, described method is for being used brightness for the region according to extracting moving object place, afterwards these regions are further screened, these regions and the Model for chromatic aberration of previously setting up are compared, only having in this region the part that is greater than threshold value with Model for chromatic aberration difference to be just construed to is prospect, with this, eliminates shade and the impact of illumination on moving target foreground extraction.
The present invention includes following step:
1), the foundation of background model:
Set a sequence threshold value, when the frame number of image acquisition reaches this threshold value, use the image sequence before threshold values to be averaging, take this average as according to setting up brightness and Model for chromatic aberration;
2), sport foreground is extracted:
The background model that the two field picture that use collects and step 1) obtain is compared, and when satisfying condition, thinks that this pixel is prospect, obtains a width bianry image thus;
3), background model is upgraded:
The brightness time to time change of monitoring environment, background model is and then carried out real-time update; During renewal, non-background pixel does not upgrade, step 2) bianry image that extracts is the foundation of upgrading as this step background pixel;
4), shade and illumination suppress:
The prospect after morphology operations that background is extracted is added up, and records the boundary value of each independent foreground area; By the result to after background extracting, carry out UNICOM's regional analysis, carry out the boundary value of fast search independence foreground area, with this, suppress shade and illumination;
5), Model for chromatic aberration upgrades:
In former having powerful connections, having new object to come into view long-time motionlessly will become new background, and the Model for chromatic aberration that step 4) is set up need to carry out real-time update.
The method of the invention concrete steps are:
Step 1: the foundation of background model:
First the foundation of background model need to set a sequence threshold value, when the frame number of image acquisition reaches this threshold value, uses image sequence to be above averaging, and take this average as according to setting up brightness and Model for chromatic aberration.As shown in Equation (1):
（1）
Wherein,
, m and n are the wide and high of image, N is sequence threshold value.When
while being luminance signal, the model of foundation is Intensity model, when
while being colour difference signal, the model of foundation is Model for chromatic aberration.With a global threshold, carry out initialization Tobin's mean variance model.
In subsequent calculations, can frequently use decimal, and the DSP that realizes the DM64x+ series of algorithm of the present invention is fixed DSP, so when setting up background model, use fractional fixed point form.
Step 2: sport foreground is extracted:
The background model that the two field picture that use collects and step 1 obtain is compared, and when meeting formula (2), thinks that this pixel is prospect.
（2）
Wherein
brightness background model,
tobin's mean variance model while being initialization,
it is the Tobin's mean variance model after model modification.Because the result obtaining is a width bianry image, in order to save space, with a pixel of each bit representative image.
Step 3: background model is upgraded:
The real-time update of background, comprises the renewal of background model and Tobin's mean variance model.Background real-time update is according to formula (3).
（3）
Wherein,
brightness background model,
it is the background model after model modification.
tobin's mean variance model while being initialization,
the Tobin's mean variance model after model modification,
a little
the pixel value at place,
it is the weight coefficient of renewal process.
Step 4: shade and illumination suppress:
The single-chip microcomputer that the present invention realizes extraction algorithm is DSP, due to the core frequency of the DSP CPU lower than PC, therefore need to suitably improve classical statistics background extracting, to meet the demand of DSP platform.Classical statistics background extracting method has been carried out to following improvement: the prospect after morphology operations that background is extracted is added up, and records the boundary value of each independent foreground area.By the result to after background extracting, carry out UNICOM's domain analysis, carry out the boundary value of fast search independence foreground area.UNICOM's domain analysis need to be passed through following step:
A). by a two field picture from top to bottom, from left to right do twice sweep.In scanning for the first time, suppose that first the non-background dot scanning is A (i, j), check so its left side A (i-1, j) and two adjacent pixels points of top A (i, j-1).
B) if. A (i-1, j) and A (i, j-1) are not labeled, and distribute a new marker character to A (i, j);
C) if. A (i-1, j) and A (i, j-1) have one to be labeled, and give the same marker character of A (i, j);
D) if. A (i-1, j) and A (i, j-1) be all labeled, so: if two marker characters are identical, give the same marker character of A (i, j), if two marker character differences, this A (i, j) is labeled as to one of them marker character, writes down two marker character equivalences simultaneously.
E). with minimum mark symbol in table of equal value, replace each marker character in table of equal value, thereby the pixel that is marked as different marker characters to belonging to same UNICOM region makes marks again.
At this moment search for each independent connected region, determine the border up and down of independent connected region, after obtaining the border of these prospects, just can determine the rectangle corresponding with each prospect.According to colour difference signal, each rectangle being carried out to background for the second time again extracts.
The square boundary providing according to connected region extracts the colour difference signal in this rectangle from a two field picture, comprises Cb and Cr signal.By the Cb in rectangle and Cr colour difference signal respectively at subtracting each other with Cb and Cr aberration background model.And with the comparison of aberration Tobin's mean variance model.If the colour difference signal of this rectangle interior pixels point meets formula (4), think that these pixels are prospects of colour difference signal.
（4）
Wherein,
the colour difference signal of a new two field picture,
aberration background model,
real-time update aberration Tobin's mean variance model afterwards,
it is initial aberration Tobin's mean variance model.
Step 5: Model for chromatic aberration upgrades:
Consider and may have object to come into view long-time motionless and become new background, the Model for chromatic aberration of foundation is also wanted real-time update, and formula (5) is followed in the renewal of Model for chromatic aberration.
（5）
Wherein,
initial aberration background model,
the aberration background model after real-time update,
the colour difference signal of a new two field picture,
initial aberration Tobin's mean variance model,
it is the aberration Tobin's mean variance model after real-time update.
The result that formula (2) and formula (4) are extracted is integrated, and those pixels that not only met formula (2) but also met formula (4) are real sport foregrounds.
Further, the bianry image obtaining after step 5 is integrated carries out morphology operations again, can filling cavity, and remove noise.
Beneficial effect of the present invention and advantage are, for moving object in site environment, be subject to ectocine brightness to change violent feature, proposed improved statistics background extracting method, the method, on original classical theory of statistics background extracting basis, is used two colour difference signals to set up Model for chromatic aberration.Due to object by illumination after, changing is colourity the most significantly, now colourity is compared brightness and more can be reflected the feature of moving object.Use brightness for the region according to extracting moving object place, afterwards these regions are further screened, these regions and the Model for chromatic aberration of previously setting up are compared, only having in this region the part that is greater than threshold value with Model for chromatic aberration difference to be construed to is prospect, has so just eliminated shade and the illumination impact on moving target foreground extraction.The present invention has good inhibiting effect to shade and illumination.
Accompanying drawing explanation
Below in conjunction with accompanying drawing, the present invention will be further described.
Fig. 1 is the moving target foreground extraction algorithm flow chart that the present invention is based on aberration;
Fig. 2 is the design sketch that classical theory of statistics background extracting of the present invention obtains;
Fig. 3 is that the present invention extracts on basis in classical theory of statistics background, the design sketch that sport foreground is further extracted according to colour difference signal.
Embodiment
below in conjunction with drawings and the specific embodiments, the present invention will be further described in detail.
Step 1: the foundation of background model
First the foundation of background model need to set a sequence threshold value, when the frame number of image acquisition reaches this threshold value, uses image sequence to be above averaging, and take this average as according to setting up brightness and Model for chromatic aberration.As shown in Equation (1):
（1）
Wherein,
, m and n are the wide and high of image, N is sequence threshold value.When
while being luminance signal, the model of foundation is Intensity model, when
while being colour difference signal, the model of foundation is Model for chromatic aberration.With a global threshold, carry out initialization Tobin's mean variance model.
In subsequent calculations, can frequently use decimal, and the DSP that realizes the DM64x+ series of algorithm of the present invention is fixed DSP, so when setting up background model, use fractional fixed point form.
Step 2: sport foreground is extracted
The background model that the two field picture that use collects and step 1 obtain is compared, and when meeting formula (2), thinks that this pixel is prospect.
（2）
Wherein
brightness background model,
tobin's mean variance model while being initialization,
it is the Tobin's mean variance model after model modification.Because the result obtaining is a width bianry image, in order to save space, with a pixel of each bit representative image.
Step 3: background model is upgraded
The real-time update of background, comprises the renewal of background model and Tobin's mean variance model.Background real-time update is according to formula (3).
（3）
Wherein,
brightness background model,
it is the background model after model modification.
tobin's mean variance model while being initialization,
the Tobin's mean variance model after model modification,
a little
the pixel value at place,
it is the weight coefficient of renewal process.
Step 4: shade and illumination suppress
The single-chip microcomputer that the present invention realizes extraction algorithm is DSP, due to the core frequency of the DSP CPU lower than PC, therefore need to suitably improve classical statistics background extracting, to meet the demand of DSP platform.Classical statistics background extracting method has been carried out to following improvement: the prospect after morphology operations that background is extracted is added up, and records the boundary value of each independent foreground area.By the result to after background extracting, carry out UNICOM's domain analysis, carry out the boundary value of fast search independence foreground area.UNICOM's domain analysis need to be passed through following step:
A). by a two field picture from top to bottom, from left to right do twice sweep.In scanning for the first time, suppose that first the non-background dot scanning is A (i, j), check so its left side A (i-1, j) and two adjacent pixels points of top A (i, j-1).
B) if. A (i-1, j) and A (i, j-1) are not labeled, and distribute a new marker character to A (i, j);
C) if. A (i-1, j) and A (i, j-1) have one to be labeled, and give the same marker character of A (i, j);
D) if. A (i-1, j) and A (i, j-1) be all labeled, so: if two marker characters are identical, give the same marker character of A (i, j), if two marker character differences, this A (i, j) is labeled as to one of them marker character, writes down two marker character equivalences simultaneously.
E). with minimum mark symbol in table of equal value, replace each marker character in table of equal value, thereby the pixel that is marked as different marker characters to belonging to same UNICOM region makes marks again.
At this moment search for each independent connected region, determine the border up and down of independent connected region, after obtaining the border of these prospects, just can determine the rectangle corresponding with each prospect.According to colour difference signal, each rectangle being carried out to background for the second time again extracts.
The square boundary providing according to connected region extracts the colour difference signal in this rectangle from a two field picture, comprises Cb and Cr signal.By the Cb in rectangle and Cr colour difference signal respectively at subtracting each other with Cb and Cr aberration background model.And with the comparison of aberration Tobin's mean variance model.If the colour difference signal of this rectangle interior pixels point meets formula (4), think that these pixels are prospects of colour difference signal.
（4）
Wherein,
the colour difference signal of a new two field picture,
aberration background model,
real-time update aberration Tobin's mean variance model afterwards,
it is initial aberration Tobin's mean variance model.
Step 5: Model for chromatic aberration upgrades
Consider and may have object to come into view long-time motionless and become new background, the Model for chromatic aberration of foundation is also wanted real-time update, and formula (5) is followed in the renewal of Model for chromatic aberration.
（5）
Wherein,
initial aberration background model,
the aberration background model after real-time update,
the colour difference signal of a new two field picture,
initial aberration Tobin's mean variance model,
it is the aberration Tobin's mean variance model after real-time update.
The result that formula (2) and formula (4) are extracted is integrated, those pixels that not only met formula (2) but also met formula (4) are real sport foregrounds, finally the bianry image obtaining after integrating is carried out to morphology operations, can filling cavity, and remove noise.
Finally it should be noted that: obviously, above-described embodiment is only for example of the present invention is clearly described, and the not restriction to embodiment.For those of ordinary skill in the field, can also make other changes in different forms on the basis of the above description.Here exhaustive without also giving all embodiments.And the apparent variation of being amplified out thus or change are still among protection scope of the present invention.
Claims (4)
1. the moving target foreground extracting method based on aberration, it is characterized in that, described method is for being used brightness for the region according to extracting moving object place, afterwards these regions are further screened, these regions and the Model for chromatic aberration of previously setting up are compared, only having in this region the part that is greater than threshold value with Model for chromatic aberration difference to be just construed to is prospect, with this, eliminates shade and the impact of illumination on moving target foreground extraction.
2. the moving target foreground extracting method based on aberration as claimed in claim 1, is characterized in that, comprises following step:
1), the foundation of background model:
Set a sequence threshold value, when the frame number of image acquisition reaches this threshold value, use the image sequence before threshold values to be averaging, take this average as according to setting up brightness and Model for chromatic aberration;
2), sport foreground is extracted:
The background model that the two field picture that use collects and step 1) obtain is compared, and when satisfying condition, thinks that this pixel is prospect, obtains a width bianry image thus;
3), background model is upgraded:
The brightness time to time change of monitoring environment, background model is and then carried out real-time update; During renewal, non-background pixel does not upgrade, step 2) bianry image that extracts is the foundation of upgrading as this step background pixel;
4), shade and illumination suppress:
The prospect after morphology operations that background is extracted is added up, and records the boundary value of each independent foreground area; By the result to after background extracting, carry out UNICOM's regional analysis, carry out the boundary value of fast search independence foreground area, with this, suppress shade and illumination;
5), Model for chromatic aberration upgrades:
In former having powerful connections, having new object to come into view long-time motionlessly will become new background, and the Model for chromatic aberration that step 4) is set up need to carry out real-time update.
3. the moving target foreground extracting method based on aberration as claimed in claim 2, is characterized in that, described method concrete steps are:
Step 1: the foundation of background model:
First the foundation of background model need to set a sequence threshold value, when the frame number of image acquisition reaches this threshold value, uses image sequence to be above averaging, and take this average as according to setting up brightness and Model for chromatic aberration; As shown in Equation (1):
（1）
Wherein,
, m and n are the wide and high of image, N is sequence threshold value;
When
while being luminance signal, the model of foundation is Intensity model, when
while being colour difference signal, the model of foundation is Model for chromatic aberration;
With a global threshold, carry out initialization Tobin's mean variance model;
In subsequent calculations, can frequently use decimal, and the DSP that realizes the DM64x+ series of algorithm of the present invention is fixed DSP, so when setting up background model, use fractional fixed point form;
Step 2: sport foreground is extracted:
The background model that the two field picture that use collects and step 1 obtain is compared, and when meeting formula (2), thinks that this pixel is prospect;
（2）
Wherein
brightness background model,
tobin's mean variance model while being initialization,
it is the Tobin's mean variance model after model modification;
Because the result obtaining is a width bianry image, in order to save space, with a pixel of each bit representative image;
Step 3: background model is upgraded:
The real-time update of background, comprises the renewal of background model and Tobin's mean variance model;
Background real-time update is according to formula (3);
（3）
Wherein,
brightness background model,
it is the background model after model modification;
tobin's mean variance model while being initialization,
the Tobin's mean variance model after model modification,
a little
the pixel value at place,
it is the weight coefficient of renewal process;
Step 4: shade and illumination suppress:
The single-chip microcomputer that the present invention realizes extraction algorithm is DSP, due to the core frequency of the DSP CPU lower than PC, therefore need to suitably improve classical statistics background extracting, to meet the demand of DSP platform;
Classical statistics background extracting method has been carried out to following improvement: the prospect after morphology operations that background is extracted is added up, and records the boundary value of each independent foreground area;
By the result to after background extracting, carry out UNICOM's domain analysis, carry out the boundary value of fast search independence foreground area;
UNICOM's domain analysis need to be passed through following step:
A). by a two field picture from top to bottom, from left to right do twice sweep;
In scanning for the first time, suppose that first the non-background dot scanning is A (i, j), check so its left side A (i-1, j) and two adjacent pixels points of top A (i, j-1);
B) if. A (i-1, j) and A (i, j-1) are not labeled, and distribute a new marker character to A (i, j);
C) if. A (i-1, j) and A (i, j-1) have one to be labeled, and give the same marker character of A (i, j);
D) if. A (i-1, j) and A (i, j-1) be all labeled, so: if two marker characters are identical, give the same marker character of A (i, j), if two marker character differences, this A (i, j) is labeled as to one of them marker character, writes down two marker character equivalences simultaneously;
E). with minimum mark symbol in table of equal value, replace each marker character in table of equal value, thereby the pixel that is marked as different marker characters to belonging to same UNICOM region makes marks again;
At this moment search for each independent connected region, determine the border up and down of independent connected region, after obtaining the border of these prospects, just can determine the rectangle corresponding with each prospect;
According to colour difference signal, each rectangle being carried out to background for the second time again extracts;
The square boundary providing according to connected region extracts the colour difference signal in this rectangle from a two field picture, comprises Cb and Cr signal;
By the Cb in rectangle and Cr colour difference signal respectively at subtracting each other with Cb and Cr aberration background model;
And with the comparison of aberration Tobin's mean variance model;
If the colour difference signal of this rectangle interior pixels point meets formula (4), think that these pixels are prospects of colour difference signal;
（4）
Wherein,
the colour difference signal of a new two field picture,
aberration background model,
real-time update aberration Tobin's mean variance model afterwards,
it is initial aberration Tobin's mean variance model;
Step 5: Model for chromatic aberration upgrades:
Consider and may have object to come into view long-time motionless and become new background, the Model for chromatic aberration of foundation is also wanted real-time update, and formula (5) is followed in the renewal of Model for chromatic aberration;
（5）
Wherein,
initial aberration background model,
the aberration background model after real-time update,
the colour difference signal of a new two field picture,
initial aberration Tobin's mean variance model,
it is the aberration Tobin's mean variance model after real-time update;
The result that formula (2) and formula (4) are extracted is integrated, and those pixels that not only met formula (2) but also met formula (4) are real sport foregrounds.
4. the moving target foreground extracting method based on aberration as claimed in claim 2 or claim 3, is characterized in that: the bianry image obtaining after step 5 is integrated carries out morphology operations again, can filling cavity, and remove noise.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410196311.XA CN103971368B (en) | 2014-05-12 | 2014-05-12 | A kind of moving target foreground extracting method based on aberration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410196311.XA CN103971368B (en) | 2014-05-12 | 2014-05-12 | A kind of moving target foreground extracting method based on aberration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103971368A true CN103971368A (en) | 2014-08-06 |
| CN103971368B CN103971368B (en) | 2017-03-15 |
Family
ID=51240817
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410196311.XA Active CN103971368B (en) | 2014-05-12 | 2014-05-12 | A kind of moving target foreground extracting method based on aberration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103971368B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993778A (en) * | 2019-04-11 | 2019-07-09 | 浙江立元通信技术股份有限公司 | A kind of method and device of determining target position |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101236606A (en) * | 2008-03-07 | 2008-08-06 | 北京中星微电子有限公司 | Shadow cancelling method and system in vision frequency monitoring |
| CN102054270A (en) * | 2009-11-10 | 2011-05-11 | 华为技术有限公司 | Method and device for extracting foreground from video image |
| US20110205236A1 (en) * | 2010-02-22 | 2011-08-25 | Casio Computer Co., Ltd. | Image processing apparatus and storage medium having stored therein an image processing program |
| CN103679704A (en) * | 2013-11-22 | 2014-03-26 | 中国人民解放军第二炮兵工程大学 | Video motion shadow detecting method based on lighting compensation |
-
2014
- 2014-05-12 CN CN201410196311.XA patent/CN103971368B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101236606A (en) * | 2008-03-07 | 2008-08-06 | 北京中星微电子有限公司 | Shadow cancelling method and system in vision frequency monitoring |
| CN102054270A (en) * | 2009-11-10 | 2011-05-11 | 华为技术有限公司 | Method and device for extracting foreground from video image |
| US20110205236A1 (en) * | 2010-02-22 | 2011-08-25 | Casio Computer Co., Ltd. | Image processing apparatus and storage medium having stored therein an image processing program |
| CN103679704A (en) * | 2013-11-22 | 2014-03-26 | 中国人民解放军第二炮兵工程大学 | Video motion shadow detecting method based on lighting compensation |
Non-Patent Citations (2)
| Title |
|---|
| 李光才: "基于视频监控的运动车辆流检测方法的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 秦秀丽: "基于YUV颜色空间和图论切割的阴影去除算法", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993778A (en) * | 2019-04-11 | 2019-07-09 | 浙江立元通信技术股份有限公司 | A kind of method and device of determining target position |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103971368B (en) | 2017-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2018086299A1 (en) | Image processing-based insulator defect detection method and system | |
| Zhu et al. | A modified neighborhood similar pixel interpolator approach for removing thick clouds in Landsat images | |
| CN105005992B (en) | A kind of based on the background modeling of depth map and the method for foreground extraction | |
| CN101236606B (en) | Shadow cancelling method and system in vision frequency monitoring | |
| CN106683112B (en) | A kind of Road domain building change detection method based on high-definition picture | |
| CN102968782B (en) | In a kind of coloured image, remarkable object takes method automatically | |
| CN101777180B (en) | Complex background real-time alternating method based on background modeling and energy minimization | |
| CN103729858B (en) | A kind of video monitoring system is left over the detection method of article | |
| CN103974053B (en) | A kind of Automatic white balance antidote extracted based on ash point | |
| CN103258332A (en) | Moving object detection method resisting illumination variation | |
| CN108986116B (en) | Mangrove forest extraction method and system based on remote sensing image | |
| CN103940824A (en) | Air electric transmission line insulator detecting method | |
| CN105160297A (en) | Masked man event automatic detection method based on skin color characteristics | |
| CN104021527B (en) | Rain and snow removal method in image | |
| WO2012005461A2 (en) | Method for automatically calculating information on clouds | |
| CN110782409B (en) | Method for removing shadow of multiple moving objects | |
| CN106157323A (en) | The insulator division and extracting method that a kind of dynamic division threshold value and block search combine | |
| CN103093471A (en) | Foreground extraction method under complicated background | |
| CN103473547A (en) | Vehicle target recognizing algorithm used for intelligent traffic detecting system | |
| CN105184812A (en) | Target tracking-based pedestrian loitering detection algorithm | |
| CN108961230A (en) | The identification and extracting method of body structure surface FRACTURE CHARACTERISTICS | |
| CN102930524A (en) | Method for detecting heads based on vertically-placed depth cameras | |
| CN106960445A (en) | A kind of cloud motion vector calculating method based on pyramid light stream | |
| CN103500451B (en) | A kind of independent floating ice extracting method for satellite data | |
| CN110659546A (en) | Illegal booth detection method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information |
Inventor after: Sun Caiying Inventor after: Li Shaobo Inventor after: New year Inventor after: Zhang Yong Inventor after: Yang Peihong Inventor before: Sun Caiying Inventor before: Lan Xiaowen Inventor before: Dong Daming |
|
| COR | Change of bibliographic data | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231229 Address after: 301500 Tianjin Binhai New Area Economic and Technological Development Zone Binhai Zhongguancun Science and Technology Park Datang Base East Area Building 2 Unit 7 301-19 Patentee after: Core Universe (Tianjin) Technology Co.,Ltd. Address before: 014010 No.7 Alwen Street, Kunqu District, Baotou City, Inner Mongolia Autonomous Region Patentee before: INNER MONGOLIA University OF SCIENCE AND TECHNOLOGY |