CN101431664A - Automatic detection method and system for intensity of passenger flow based on video image - Google Patents
Automatic detection method and system for intensity of passenger flow based on video image Download PDFInfo
- Publication number
- CN101431664A CN101431664A CNA2007100478386A CN200710047838A CN101431664A CN 101431664 A CN101431664 A CN 101431664A CN A2007100478386 A CNA2007100478386 A CN A2007100478386A CN 200710047838 A CN200710047838 A CN 200710047838A CN 101431664 A CN101431664 A CN 101431664A
- Authority
- CN
- China
- Prior art keywords
- crowd
- image
- video image
- passenger flow
- video
- 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.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method for checking dense passenger flow intensity based on video image which belongs to a computer visual technique field. The invention uses a video collecting device and a processing algorithm. A video image collecting device collects monitor video image through CCTV, usually collects video image of passenger out-in flow by using a camera put on top of out-in port in passenger flow chunnel in time. The processor processes collected video image using computer visual algorithm, when passenger flow intensity is low, pixels statistics with less algorithm amount but veracious is used, and information of time shaft is used for generating background; when passenger flow intensity is high, passenger image is processed multi-angle analysis using wavelet packets decomposition, then box-counting dimension of wavelet packets decomposition coefficient matrix is extracted as characteristic for sending to classifier and obtaining passenger intensity grade, reaches aims at checking dense passenger flow intensity in real time.
Description
Technical field
The present invention relates to computer vision field, specifically be based on the intensity of passenger flow automatic testing method and the device of video image, can be widely used in the bigger public administration fields of the volume of the flow of passengers such as airport, square, subway, fire (visitor) station, market.
Background technology
Along with rapid development of economy, the urbanization degree of population is more and more higher, and the density of population in city is increasing, and the crowd's problem of management in some public place becomes increasingly conspicuous.Crowd density is an important references index that characterizes the instant degree of crowding of particular place, it is the important evidence that the public place is effectively managed, along with the fast development of domestic economy and the transportation of extensive passenger flow, at some particular place the potential demand of crowd density has been become more and more urgent at present.The conventional method that crowd density is estimated is artificial estimation, but this method is relatively more subjective, can not do quantitative judgement.The automated process of crowd density estimation recently also grows up gradually.But these methods respectively have deficiency, and when crowd density was higher, because crowd's overlapping phenomenon, the linear relationship between measured value and the crowd's number disappeared, and causes error very big, and these methods require to provide the background image of scene.Though can solve high density crowd's estimation problem, amount of calculation is bigger, the processing time is longer, and this method does not consider that the video camera perspective effect causes the big problem of error.
Avoiding one of method of occlusion issue, is by adjusting the shooting angle of video camera, adopting the way of taking from people's head downwards.The mode of this shooting has reduced the difficulty of handling undoubtedly, but need reinstall the video camera of detection, has increased the input of system, and in addition, sometimes supervisory control system needs the information of acquisitor's face.Shi Yong mode is more, and detection is as the enhancement function of safety monitoring system (closed-circuit television CCTV) automatically with intensity of passenger flow, and the treatment technology by video image obtains intensity of passenger flow.
Summary of the invention
The objective of the invention is at the deficiencies in the prior art, propose a kind of intensity of passenger flow automatic testing method based on video image.In actual intensive duty traffic amount detected, this method solved at all that the error that exists in the detection method in the past is big, the processing time is long, video camera transmission effect considered problem such as deficiency.From the collection of video image, the Treatment Analysis that is edited into data and result's output, have and calculate easy, real-time, good reliability, be easy to transplant, be convenient to advantages such as expansion and maintenance.
A kind of intensity of passenger flow automatic testing method based on video image, step is as follows;
(a) gather the monitor video image that passenger flow is come in and gone out, obtain crowd's image;
(b) background process shields the crowd's image that collects through template, remove the complex background that does not comprise people's group object, keeps the target area that comprises people's group object, carries out the medium filtering denoising then;
(c) adopt computer vision algorithms make to the video image that the collects processing of classifying, finally obtain the crowd density grade,
(i) when crowd density is low, adopt the less but pixel count statistics more accurately of amount of calculation, and the information of using time shaft carries out background and generates,
(ii) when crowd density is higher, adopt the texture analysis method to calculate crowd density.
As above-mentioned method wherein, the described step (c-ii) that crowd density classification in the video image is handled comprising: at first crowd's image carries out WAVELET PACKET DECOMPOSITION, the meter box counting dimension that extracts each sub-band images then adopts grader to obtain crowd density grade result as tagsort.
As above-mentioned method wherein, be characterised in that described step (c-i) to crowd density classification in the video image comprising:
1) adopt threshold method to obtain the bianry image of crowd's prospect, bianry image is to judge whether pixel is the threshold value of crowd's prospect, and this threshold value is by concrete measuring, can follow the variation of basic data in the background process and adjusts;
2) perspective effect of crowd's prospect is proofreaied and correct, region of interest ROI be divided into that area equates far away, in, nearly three zones, calculate each regional crowd density respectively, get trizonal crowd density average then as the final crowd density of estimating.
As above-mentioned method wherein, be characterised in that in the step (b) crowd's picture being handled and need be carried out multiscale analysis to judging crowd's prospect, it is as follows that the video camera perspective effect is carried out treatment step:
(a) image is carried out orthogonal wavelet and decompose the sub-band images that obtains on the different scale direction;
(b) use WAVELET PACKET DECOMPOSITION the sub-band images that step (a) obtains is carried out multiscale analysis;
(c) subband of using all different directions different scales carries out texture analysis, calculates several box counting dimensions;
(d) utilize the texture pattern of crowd's image that tangible self-similarity and fractal characteristic are arranged, use fractal dimension and characterize the characteristic vector that texture pattern obtains reflecting dimensional properties, respectively yardsticks at different levels are extracted characteristic parameter.
As above-mentioned method wherein, in order to obtain to reflect the characteristic vector of dimensional properties, need respectively yardsticks at different levels to be extracted characteristic parameter in the step (d); Texture pattern according to crowd's image has tangible self-similarity and fractal characteristic, adopts fractal dimension to characterize this texture pattern, adopts meter box method to calculate fractal dimension, and dimension is as characteristic vector.
As above-mentioned method specifically is Support Vector Machine in step (c-ii) to grader in the crowd density classification wherein, in multiclass is divided, need form tree with Support Vector Machine classifiers group and divide.
Realize the claim system for carrying out said process, be characterised in that to comprise:
(a) video image acquisition equipment comprises the video camera on the top, gateway that places the passenger flow passage, closed-circuit television system CCTV;
(b) video process apparatus of the video image that collects being handled with computer vision algorithms make specifically comprises video acquisition module, memory, processor, and described processor includes but not limited to flush bonding processor; Platform is made up of TMS320C64x and TMS320C62x fixed point series and TMS320C67x floating-point series.
(c) the meter box counting dimension result of WAVELET PACKET DECOMPOSITION coefficient matrix is carried out the grader of analyzing and processing, grader adopts Support Vector Machine and support vector unit.
The present invention is achieved by the following technical solutions, and the present invention adopts video acquisition device and Processing Algorithm, realizes the real-time detection to the intensive duty traffic metric density.Wherein video image acquisition equipment system (CCTV) acquisition monitoring video image by way of closed-circuit television adopts the video camera on the top, gateway that places the passenger flow passage usually, gathers the video image that passenger flow is come in and gone out in real time.Processor adopting video image Processing Algorithm is handled the video image that collects, and when crowd density is low, adopts the less but pixel count statistics more accurately of amount of calculation, and the information of using time shaft is carried out background and generated; Using WAVELET PACKET DECOMPOSITION when crowd density is higher comes crowd's image is carried out multi-angular analysis, the meter box counting dimension that extracts the WAVELET PACKET DECOMPOSITION coefficient matrix then is as feature, send into grader and classify, obtain the crowd density grade, reach the real-time testing goal of intensity of passenger flow.
The inventive method comprises following step:
1, intensive duty traffic video acquisition
On-the-spot in intensive duty traffic monitoring on a large scale, can carry out the real-time imaging record to intensive duty traffic by monitoring camera is installed, by video camera obtain be the simulation vision signal, the vision signal of these simulations to be handed over Computer Processing is arranged, and on digital line, transmit, must adopt corresponding apparatus that analog signal is converted into digital signal and compress.The present invention adopts high performance video acquisition compressing card to finish collection of high density passenger flow live video and compression, analog video signal real-time digitization compressed encoding can be transferred to nucleus module then and directly handled by it.
2, video image is carried out based on multiscale analysis and fractal texture analysis
Decompose if piece image is carried out orthogonal wavelet, just can obtain the sub-band images on the different scale different directions.This sub-band images has comprised the detailed information of particular dimensions and direction, has also comprised spatial information (si), can rebuild original image fully according to these subbands.Yet the orthogonal wavelet decomposition is only carried out higher one-level decomposition to low frequency sub-band, and in order to obtain the multiscale analysis of high-frequency sub-band, the present invention uses WAVELET PACKET DECOMPOSITION crowd's image is carried out multiscale analysis.Because sub-band images is to obtain by the wavelet transform down-sampling, the texture structure information that two relative distance pixels far away are represented can be represented by two nearer pixels of relative distance on senior sub-band images, carry out texture analysis so the present invention uses the subband of all different directions different scales.In order to access the characteristic vector of reflection dimensional properties, the present invention extracts characteristic parameter to yardsticks at different levels respectively.Because the texture pattern of crowd's image has tangible self-similarity and fractal characteristic, so the present invention characterizes texture pattern with fractal dimension.
3, the utilization Support Vector Machine is carried out grade separation to crowd density
Support Vector Machine (SVM) is a kind of new general mode identification method that grows up on the Statistical Learning Theory basis; At present, the result in two class partition problems is better for Support Vector Machine, in multiclass is divided, need form tree with the support vector unit and divide.Consider that intensity of passenger flow is to carry out respective classified on a plurality of orders of density, what the present invention adopted is that the tree that the support vector unit is formed comes final crowd density is carried out grade separation.
The present invention is made of video acquisition module, memory, processor, and described processor is a flush bonding processor.The present invention can effectively solve a difficult problem that exists in the high density passenger flow statistics, particularly at large size city, and the traffic places with dense, intensity of passenger flow is very high, has the serious phenomenon of blocking, and can not effectively solve by traditional method of video image processing.The present invention is fully based on existing closed-circuit television system, in conjunction with computer vision means and image processing algorithm, detect intensity of passenger flow, the density information of acquisition can be used as the foundation of public people's flow management and the design of passenger flow layout of roads, and the foundation of passenger flow channel management.
Description of drawings
Fig. 1 is based on the intensity of passenger flow automatic checkout system structured flowchart of video image;
Fig. 2 is an intensity of passenger flow detection method software flow pattern of the present invention;
Fig. 3 is a software test of the present invention interface schematic diagram.
Embodiment
Below in conjunction with accompanying drawing embodiments of the invention are elaborated.
Present embodiment adopts the automatic detection scheme of the intensity of passenger flow based on video image shown in Figure 1, and concrete implementation step is as follows:
1, the foundation of hardware device
As shown in Figure 1, can adopt with the digital signal processor DSP based on the automatic detection of intensive duty traffic metric density of video image is the hardware configuration of core, is made of acquisition module, memory, processor.Because computer vision algorithms make has certain complexity, processing to as if the complex video image, and require system can satisfy real-time, therefore the chip of selecting must have powerful disposal ability.Comprehensive above each side is considered, can select the TMS320DM642 of TI.The TMS320C6000DSP platform has improved performance and cost-benefit level, provides industry the fastest extensive DSP product line, and these DSP are with the clock speed operation up to 1GHz.Platform is made up of TMS320C64x and TMS320C62x fixed point series and TMS320C67x floating-point series.C6000 DSP platform is the designer's of products such as processing target broadband infrastructure, high-performance audio frequency and imaging applications optimal selection.TMS320DM642 is the main product in the multimedia process field that TI company releases at present, and it is on the basis of C64x, has increased many ancillary equipment and interface.
2, the foundation of software flow pattern
As shown in Figure 2, intensity of passenger flow detection method software flow pattern is at first gathered video image, and crowd's image of input at first masks the complex background that does not comprise people's group object through template, only keep the region of interest that comprises people's group object, carry out the medium filtering denoising then.Background subtracts the crowd image subtracting background reference picture of computing after with medium filtering, obtains the foreground image of people's group object, calculates the ratio that crowd's foreground image area accounts for the region of interest area, promptly is the reference value of crowd density after Geometric corrections.Because there is the overlapping phenomenon in the crowd, must estimate crowd density with the method for texture analysis.The texture analysis method is at first carried out WAVELET PACKET DECOMPOSITION to the crowd's image behind the medium filtering, extracts the meter box counting dimension of each sub-band images then and sends into the Support Vector Machine classification as feature, obtains the crowd density grade.Under low-density crowd situation, crowd density is expressed as the pixel count of crowd's prospect and the ratio of the pixel count of region of interest.The present invention adopts threshold method to obtain the bianry image of crowd's prospect.Judge that whether pixel is that the threshold value of crowd's prospect is determined by experiment.Because the optical axis and the horizontal plane of video camera acutangulate, there is near big and far smaller perspective effect in crowd's image of shooting, can produce certain influence to the crowd density estimated result like this, brings error.For perspective effect is proofreaied and correct, ROI be divided into that area equates far away, in, nearly three zones, calculate each regional crowd density respectively, get trizonal crowd density average then as the final crowd density of estimating.
3, the intensive duty traffic image is set up multiscale analysis
Decompose if piece image is carried out orthogonal wavelet, just can obtain the sub-band images on the different scale different directions.This sub-band images has comprised the detailed information of particular dimensions and direction, has also comprised spatial information (si), can rebuild original image fully according to these subbands.Yet the orthogonal wavelet decomposition is only carried out higher one-level decomposition to low frequency sub-band, and in order to obtain the multiscale analysis of high-frequency sub-band, the present invention uses WAVELET PACKET DECOMPOSITION crowd's image is carried out multiscale analysis.Because sub-band images is to obtain by the wavelet transform down-sampling, the texture structure information that two relative distance pixels far away are represented can be represented by two nearer pixels of relative distance on senior sub-band images, carry out texture analysis so the present invention uses the subband of all different directions different scales.In order to access the characteristic vector of reflection dimensional properties, the present invention extracts characteristic parameter to yardsticks at different levels respectively.Because the texture pattern of crowd's image has tangible self-similarity and fractal characteristic, so the present invention characterizes texture pattern with fractal dimension.
4, the extraction of characteristic vector and classification
In order to obtain to reflect the characteristic vector of dimensional properties, need respectively yardsticks at different levels to be extracted characteristic parameter.Because the texture pattern of crowd's image has tangible self-similarity and fractal characteristic, so the present invention characterizes this texture pattern with fractal dimension.Here adopt meter box method to calculate fractal dimension.
It is exactly method with similar expansion that meter box method is calculated fractal dimension, is 1-15 structural element with size herein, and the method derivation fitting function of employing least mean-square estimate.
The slope of fitting function=(mxy-mx*my)/(mxx-mx*mx);
The slope of Fractal Dimension=2-fitting function
Parameter is as follows:
mx=(log(3)+log(5)+log(7)+log(9)+log(11)+log(13))/7;
my=(log(dense1)+log(dense3)+log(dense5)+log(dense7)+log(dense9)+log(dense11)+
log(dense13))/7;
mxy=(log(1)*log(dense1)+log(3)*log(dense3)+log(5)*log(dense5)+log(7)*log(dense
7)+log(9)*log(dense9)+log(11)*log(dense11)+log(13)*log(dense13))/7;
mxx=(log(3)*log(3)+log(5)*log(5)+log(7)*log(7)+log(9)*log(9)+log(11)*log(11)+lo
g(13)*log(13))/7;
Sorting technique is because crowd's image training sample is less, and Support Vector Machine has distinctive advantage on the small sample pattern recognition problem, so the present invention adopts Support Vector Machine as grader.At present, the result in two class partition problems is better for Support Vector Machine, because the present invention is a multiclass partition problem, adopts the support vector unit to form tree here and divides.
5. test result
Software test interface schematic diagram as shown in Figure 3.
Present classification accuracy is about 70%.
Claims (7)
1. intensity of passenger flow automatic testing method based on video image, step is as follows;
(a) gather the monitor video image that passenger flow is come in and gone out, obtain crowd's image;
(b) background process shields the crowd's image that collects through template, remove the complex background that does not comprise people's group object, keeps the target area that comprises people's group object, carries out the medium filtering denoising then;
(c) adopt computer vision algorithms make to the video image that the collects processing of classifying, finally obtain the crowd density grade,
(i) when crowd density is low, adopt the less but pixel count statistics more accurately of amount of calculation, and the information of using time shaft carries out background and generates,
(ii) when crowd density is higher, adopt the texture analysis method to calculate crowd density.
2. method according to claim 1, be characterised in that, the described step (c-ii) that crowd density classification in the video image is handled comprising: at first crowd's image carries out WAVELET PACKET DECOMPOSITION, the meter box counting dimension that extracts each sub-band images then adopts grader to obtain crowd density grade result as tagsort.
3. method according to claim 1 is characterised in that, described step (c-i) to crowd density classification in the video image comprising:
1) adopt threshold method to obtain the bianry image of crowd's prospect, bianry image is to judge whether pixel is the threshold value of crowd's prospect, and this threshold value is by concrete measuring, can follow the variation of basic data in the background process and adjusts;
2) perspective effect of crowd's prospect is proofreaied and correct, region of interest ROI be divided into that area equates far away, in, nearly three zones, calculate each regional crowd density respectively, get trizonal crowd density average then as the final crowd density of estimating.
4. method according to claim 1 is characterised in that, in the step (b) crowd's picture being handled and need be carried out multiscale analysis to judging crowd's prospect, it is as follows that the video camera perspective effect is carried out treatment step:
(a) image is carried out orthogonal wavelet and decompose the sub-band images that obtains on the different scale direction;
(b) use WAVELET PACKET DECOMPOSITION the sub-band images that step (a) obtains is carried out multiscale analysis;
(c) subband of using all different directions different scales carries out texture analysis, calculates several box counting dimensions;
(d) utilize the texture pattern of crowd's image that tangible self-similarity and fractal characteristic are arranged, use fractal dimension and characterize the characteristic vector that texture pattern obtains reflecting dimensional properties, respectively yardsticks at different levels are extracted characteristic parameter.
5. method according to claim 4 is characterised in that, in order to obtain reflecting the characteristic vector of dimensional properties, needs respectively yardsticks at different levels to be extracted characteristic parameter in the step (d); Texture pattern according to crowd's image has tangible self-similarity and fractal characteristic, adopts fractal dimension to characterize this texture pattern, adopts meter box method to calculate fractal dimension, and dimension is as characteristic vector.
6. method according to claim 2 is characterised in that, grader specifically is a Support Vector Machine in step (c-ii) is classified to crowd density, in multiclass is divided, need form tree with Support Vector Machine classifiers group and divide.
7. realize the system of the described method of claim 1-6, be characterised in that to comprise:
(a) video image acquisition equipment comprises the video camera on the top, gateway that places the passenger flow passage, closed-circuit television system CCTV;
(b) video process apparatus of the video image that collects being handled with computer vision algorithms make specifically comprises video acquisition module, memory, processor, and described processor includes but not limited to flush bonding processor; Platform is made up of TMS320C64x and TMS320C62x fixed point series and TMS320C67x floating-point series.
(c) the meter box counting dimension result of WAVELET PACKET DECOMPOSITION coefficient matrix is carried out the grader of analyzing and processing, grader adopts Support Vector Machine and support vector unit.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2007100478386A CN101431664A (en) | 2007-11-06 | 2007-11-06 | Automatic detection method and system for intensity of passenger flow based on video image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2007100478386A CN101431664A (en) | 2007-11-06 | 2007-11-06 | Automatic detection method and system for intensity of passenger flow based on video image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101431664A true CN101431664A (en) | 2009-05-13 |
Family
ID=40646791
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA2007100478386A Pending CN101431664A (en) | 2007-11-06 | 2007-11-06 | Automatic detection method and system for intensity of passenger flow based on video image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101431664A (en) |
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101835034A (en) * | 2010-05-27 | 2010-09-15 | 王巍 | Crowd characteristic counting system |
CN101959060A (en) * | 2010-10-26 | 2011-01-26 | 镇江科大船苑计算机网络工程有限公司 | Video passenger flow monitoring system and method |
CN102034226A (en) * | 2010-12-28 | 2011-04-27 | 北京智安邦科技有限公司 | Method and device for removing shading points |
CN102136141A (en) * | 2010-01-26 | 2011-07-27 | 三洋电机株式会社 | Congestion degree measuring apparatus |
CN102360450A (en) * | 2011-09-26 | 2012-02-22 | 华中科技大学 | Method for counting number of people based on masses |
CN102496058A (en) * | 2011-11-11 | 2012-06-13 | 北京声迅电子股份有限公司 | Passenger flow density detection method |
CN103049765A (en) * | 2012-12-21 | 2013-04-17 | 武汉经纬视通科技有限公司 | Method for judging crowd density and number of people based on fish eye camera |
CN102044073B (en) * | 2009-10-09 | 2013-05-29 | 汉王科技股份有限公司 | Method and system for judging crowd density in image |
CN103226860A (en) * | 2013-04-12 | 2013-07-31 | 中国民航大学 | Passage passenger traffic density estimation method |
CN103489012A (en) * | 2013-09-30 | 2014-01-01 | 深圳市捷顺科技实业股份有限公司 | Crowd density detecting method and system based on support vector machine |
CN104123840A (en) * | 2014-07-28 | 2014-10-29 | 重庆大学 | Video detection technology based traffic density qualitative judging method |
CN104268898A (en) * | 2014-09-15 | 2015-01-07 | 郑州天迈科技股份有限公司 | Method for detecting density of passengers in bus on basis of image analysis |
CN104408721A (en) * | 2014-11-25 | 2015-03-11 | 西安交通大学 | Stamper image extracting method based on background density estimation |
CN104778670A (en) * | 2015-04-17 | 2015-07-15 | 广西科技大学 | Fractal-wavelet self-adaption image denoising method based on multivariate statistical model |
CN105205900A (en) * | 2015-10-23 | 2015-12-30 | 华录智达科技有限公司 | Dynamic self-adaptive public transport passenger flow statistic device based on video recognition |
CN105740819A (en) * | 2016-01-29 | 2016-07-06 | 中国科学院信息工程研究所 | Integer programming based crowd density estimation method |
CN105791774A (en) * | 2016-03-31 | 2016-07-20 | 北京工业大学 | Surveillance video transmission method based on video content analysis |
CN104392522B (en) * | 2014-11-21 | 2016-08-24 | 安徽清新互联信息科技有限公司 | A kind of passenger flow statistical system based on pseudo-depth information |
CN106295460A (en) * | 2015-05-12 | 2017-01-04 | 株式会社理光 | The detection method of people and equipment |
CN106909787A (en) * | 2017-02-27 | 2017-06-30 | 中南林业科技大学 | A kind of spark plug gap Forecasting Methodology and device |
CN108171162A (en) * | 2017-12-27 | 2018-06-15 | 重庆交通开投科技发展有限公司 | Crowded degree detection method, apparatus and system |
CN109474889A (en) * | 2018-12-12 | 2019-03-15 | 维沃移动通信(杭州)有限公司 | Information transferring method, mobile terminal and server |
CN112001274A (en) * | 2020-08-06 | 2020-11-27 | 腾讯科技(深圳)有限公司 | Crowd density determination method, device, storage medium and processor |
-
2007
- 2007-11-06 CN CNA2007100478386A patent/CN101431664A/en active Pending
Cited By (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102044073B (en) * | 2009-10-09 | 2013-05-29 | 汉王科技股份有限公司 | Method and system for judging crowd density in image |
CN102136141A (en) * | 2010-01-26 | 2011-07-27 | 三洋电机株式会社 | Congestion degree measuring apparatus |
CN101835034A (en) * | 2010-05-27 | 2010-09-15 | 王巍 | Crowd characteristic counting system |
CN101959060A (en) * | 2010-10-26 | 2011-01-26 | 镇江科大船苑计算机网络工程有限公司 | Video passenger flow monitoring system and method |
CN101959060B (en) * | 2010-10-26 | 2012-06-06 | 镇江科大船苑计算机网络工程有限公司 | Video passenger flow monitoring system and method |
CN102034226A (en) * | 2010-12-28 | 2011-04-27 | 北京智安邦科技有限公司 | Method and device for removing shading points |
CN102034226B (en) * | 2010-12-28 | 2012-05-30 | 北京智安邦科技有限公司 | Method and device for removing shading points |
CN102360450A (en) * | 2011-09-26 | 2012-02-22 | 华中科技大学 | Method for counting number of people based on masses |
CN102496058B (en) * | 2011-11-11 | 2014-09-17 | 北京声迅电子股份有限公司 | Passenger flow density detection method |
CN102496058A (en) * | 2011-11-11 | 2012-06-13 | 北京声迅电子股份有限公司 | Passenger flow density detection method |
CN103049765A (en) * | 2012-12-21 | 2013-04-17 | 武汉经纬视通科技有限公司 | Method for judging crowd density and number of people based on fish eye camera |
CN103226860B (en) * | 2013-04-12 | 2015-05-20 | 中国民航大学 | Passage passenger traffic density estimation method |
CN103226860A (en) * | 2013-04-12 | 2013-07-31 | 中国民航大学 | Passage passenger traffic density estimation method |
CN103489012A (en) * | 2013-09-30 | 2014-01-01 | 深圳市捷顺科技实业股份有限公司 | Crowd density detecting method and system based on support vector machine |
CN103489012B (en) * | 2013-09-30 | 2017-05-24 | 深圳市捷顺科技实业股份有限公司 | Crowd density detecting method and system based on support vector machine |
CN104123840A (en) * | 2014-07-28 | 2014-10-29 | 重庆大学 | Video detection technology based traffic density qualitative judging method |
CN104123840B (en) * | 2014-07-28 | 2016-05-18 | 重庆大学 | Vehicle density qualitative discrimination method based on video detection technology |
CN104268898A (en) * | 2014-09-15 | 2015-01-07 | 郑州天迈科技股份有限公司 | Method for detecting density of passengers in bus on basis of image analysis |
CN104392522B (en) * | 2014-11-21 | 2016-08-24 | 安徽清新互联信息科技有限公司 | A kind of passenger flow statistical system based on pseudo-depth information |
CN104408721A (en) * | 2014-11-25 | 2015-03-11 | 西安交通大学 | Stamper image extracting method based on background density estimation |
CN104408721B (en) * | 2014-11-25 | 2017-02-22 | 西安交通大学 | Stamper image extracting method based on background density estimation |
CN104778670A (en) * | 2015-04-17 | 2015-07-15 | 广西科技大学 | Fractal-wavelet self-adaption image denoising method based on multivariate statistical model |
CN106295460A (en) * | 2015-05-12 | 2017-01-04 | 株式会社理光 | The detection method of people and equipment |
CN106295460B (en) * | 2015-05-12 | 2019-05-03 | 株式会社理光 | The detection method and equipment of people |
CN105205900A (en) * | 2015-10-23 | 2015-12-30 | 华录智达科技有限公司 | Dynamic self-adaptive public transport passenger flow statistic device based on video recognition |
CN105740819A (en) * | 2016-01-29 | 2016-07-06 | 中国科学院信息工程研究所 | Integer programming based crowd density estimation method |
CN105791774A (en) * | 2016-03-31 | 2016-07-20 | 北京工业大学 | Surveillance video transmission method based on video content analysis |
CN106909787A (en) * | 2017-02-27 | 2017-06-30 | 中南林业科技大学 | A kind of spark plug gap Forecasting Methodology and device |
CN108171162A (en) * | 2017-12-27 | 2018-06-15 | 重庆交通开投科技发展有限公司 | Crowded degree detection method, apparatus and system |
CN108171162B (en) * | 2017-12-27 | 2021-05-11 | 重庆交通开投科技发展有限公司 | Crowd crowding degree detection method, device and system |
CN109474889A (en) * | 2018-12-12 | 2019-03-15 | 维沃移动通信(杭州)有限公司 | Information transferring method, mobile terminal and server |
CN112001274A (en) * | 2020-08-06 | 2020-11-27 | 腾讯科技(深圳)有限公司 | Crowd density determination method, device, storage medium and processor |
CN112001274B (en) * | 2020-08-06 | 2023-11-17 | 腾讯科技(深圳)有限公司 | Crowd density determining method, device, storage medium and processor |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101431664A (en) | Automatic detection method and system for intensity of passenger flow based on video image | |
CN107330372B (en) | Analysis method of video-based crowd density and abnormal behavior detection system | |
CN103218816B (en) | A kind of crowd density estimation method and people flow rate statistical method based on video analysis | |
CN105976383A (en) | Power transmission equipment fault diagnosis method based on limit learning machine image recognition | |
CN105791774A (en) | Surveillance video transmission method based on video content analysis | |
CN104504377B (en) | A kind of passenger on public transport degree of crowding identifying system and method | |
CN110569755B (en) | Intelligent accumulated water detection method based on video | |
EP2709066A1 (en) | Concept for detecting a motion of a moving object | |
CN105551267B (en) | A kind of method that traffic signaling equipment is controlled by vehicle flowrate | |
KR102199252B1 (en) | Method and Apparatus for Analyzing Traffic Situation | |
CN111667655A (en) | Infrared image-based high-speed railway safety area intrusion alarm device and method | |
CN112464893A (en) | Congestion degree classification method in complex environment | |
Wang et al. | Crowd density estimation based on texture feature extraction | |
Ua-Areemitr et al. | Low-cost road traffic state estimation system using time-spatial image processing | |
CN103996266A (en) | Tailing reservoir safety monitoring device based on video analysis | |
EP2709065A1 (en) | Concept for counting moving objects passing a plurality of different areas within a region of interest | |
CN108198422A (en) | A kind of road ponding extraction system and method based on video image | |
CN102034243A (en) | Method and device for acquiring crowd density map from video image | |
CN116402994A (en) | Railway danger monitoring method based on laser radar and video image fusion | |
CN105426811B (en) | A kind of crowd's abnormal behaviour and crowd density recognition methods | |
Liang et al. | [Retracted] Vehicle Detection Algorithm Based on Embedded Video Image Processing in the Background of Information Technology | |
Shuyun | Real-time traffic flow detection system based on video image processing | |
CN106485281A (en) | A kind of detection method of the Air haze class of pollution | |
Tusch et al. | Efficient level of service classification for traffic monitoring in the compressed video domain | |
Lu et al. | Shadow removal based on shadow direction and shadow attributes |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Open date: 20090513 |