CN106951885A - A kind of people flow rate statistical method based on video analysis - Google Patents
A kind of people flow rate statistical method based on video analysis Download PDFInfo
- Publication number
- CN106951885A CN106951885A CN201710225976.2A CN201710225976A CN106951885A CN 106951885 A CN106951885 A CN 106951885A CN 201710225976 A CN201710225976 A CN 201710225976A CN 106951885 A CN106951885 A CN 106951885A
- Authority
- CN
- China
- Prior art keywords
- video
- pedestrian
- detection
- people
- head
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V10/507—Summing image-intensity values; Histogram projection analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30242—Counting objects in image
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a kind of people flow rate statistical method based on video analysis, at least one set of pedestrian that the video is shot using camera from top area monitors Sample video and one group of pedestrian's monitor and detection video, and the demographic method comprises the following steps：S10：The number of people image in pedestrian's monitoring Sample video and detection video is gathered, positive sample training image collection and the negative sample training image collection is built；S20：HOG is extracted respectively；S30：The off-line training of SVM heads grader is carried out to the gradient orientation histogram feature for training input picture；S40：Real-time head detection；S50：Target following is carried out using optical flow method；S60：ROI is set；S70：Statistical number of person.The inventive method can be counted to the flow of the people of specific region, and check picture not in the same time in mobile phone terminal, relative to the method for human testing, can avoid the problem of detection target is blocked mutually；Build on wide variety of video monitoring equipment, low cost, statistical accuracy are high.
Description
Technical field
The present invention relates to a kind of IMAQ analytical technology, more particularly to a kind of people flow rate statistical side based on video analysis
Method.
Background technology
In the management and decision-making of the public places such as market, airport, museum, flow of the people is indispensable data, with
Exemplified by market, flow of the people is very basic and important index, and the sale with market is closely related, by flow of the people, can be pin
Sell, service and the reliable reference information of logistics offer.In some regions, flow of the people as the control area density of population parameter,
It is a significant consideration of its security again.So, by the statistics to flow of the people, can effectively it monitor in real time, group
The operation of public place is knitted, environment and the better service of safety are provided for people.
At present, the method for obtaining pedestrian's traffic statistics can be largely classified into three major types：1st, artificial statistics or contact equipment；
2nd, pedestrian's traffic statistics are carried out using sensor；3rd, the people flow rate statistical based on computer vision.Using manually to pedestrian's flow
Counted, although precision can meet requirement, but consume manpower, financial resources, and not possess systemic and comprehensive；Rotation
The contact equipment such as door, rotating handles, although departing from artificial, but bring inconvenience, and one time one to pedestrian to force sense
As people's stream statistics of public sphere can not only be suitable for by a people；And detect what human body was counted using sensor
Method, most widely used at present is infrared detection system, is also mainly used in passage, and such system is contactless, is compared
Contact equipment is greatly improved.But in actual motion there is also it is very important the problem of, or such as using single
People's passage limit the big flow stream of people by convenience, and when using fat pipe when, multiple pedestrians are consecutively or simultaneously by red
During outer device, just occur and block, the degree of accuracy of demographics will be substantially reduced, in addition, single pedestrian is in the temporary of detection zone
When stop, or body and belongings interference, very big influence can be also brought to the accuracy of statistics.
The content of the invention
It is contemplated that at least solving technical problem present in prior art or correlation technique.
In recent years, along with the development of the correlation techniques such as computer vision, pattern-recognition and artificial intelligence, the field of application
Also gradually expanding, video monitoring system is increasingly widely used in the every aspect of people's life.To video sequence
In pedestrian extract, identification and tracking on the basis of, further carry out accurate demographics for the present invention a kind of thought.
Therefore, the invention provides a kind of people flow rate statistical method based on video analysis, the video is using taking the photograph
As at least one set of pedestrian monitoring Sample video and one group of pedestrian's monitor and detection video that head is shot from top area, the number system
Meter method comprises the following steps：
S10：The number of people image in pedestrian's monitoring Sample video is gathered, so that positive sample training image collection is constituted, with described
Those in pedestrian's monitoring Sample video in addition to number of people image are not pedestrian heads but the image conduct of doubtful pedestrian head
Negative sample training image collection, and the positive sample training image collection and the negative sample training image collection are normalized
Into the training input picture of an equal amount of size；The monitoring of statistical number of person the need for gathering in pedestrian's monitor and detection video
Area image frame；
S20：The gradient orientation histogram feature of training input picture and monitor area picture frame is extracted respectively（HOG）；Gradient side
To histogram feature（HOG）The gradient direction of mainly localized region is calculated, and is then described with histogram, i.e., in fact
A kind of feature descriptor of image local overlapping region；
S30：The off-line training of SVM heads grader is carried out to the gradient orientation histogram feature for training input picture, a large amount of
Positive sample training image collection and the negative sample training image collection after step S20 processing are input to SVMs mould
It is trained in type, obtains the optimum classifier being directed under these training samples；SVMs（SVM）It is that a kind of effort is sought
A kind of machine learning method of structural risk minimization；
S40：Real-time head detection, the monitor area picture frame of the detection video sequence after step S20 processing is input to
In the grader trained, by setting detection window moving step length and scaling the dimension scale of abstract image window, detection is allowed
Window is scanned on different positions, differentiates that these windows are that head zone is also non-head according to the result of decision of grader
Portion region, finally marks the position on head in the video sequence；
S50：Target following is carried out using optical flow method, realize the head movement in video frame image is carried out lasting detection,
With the tracking with location updating message；
S60：ROI is set, area-of-interest is realized using the setting of the ROI inside opencv, detachment's area-of-interest is carried out
Processing；Area-of-interest can arbitrarily be set according to actual requirement, so that improving system obtains practicality；
S70：Statistical number of person, according to the movement locus on head, contrast is imported and exported mark and can determine whether into outgoing direction, when moving object is first
Encounter import mark and then proceed to motion and encounter exit marking to enter behavior, when moving object first encounter exit marking then after
Reforwarding is dynamic to encounter import labeled as behavior of going out, and finally, turnover behavior is judged according to mark and head movement track is imported and exported, with
Realize the counting statistics of number.
Invention has advantages below and beneficial effect compared with prior art：
The demographic method of sequence and image processing techniques, builds on wide variety of video monitoring equipment, low cost, should
It is high with wide, statistical accuracy.
Further, also including step S80:The picture detected, is saved as img.jpg formatted files by remote inquiry,
A notice is sent out to pathon programs while preservation, picture file server is saved in, then send to remote control terminal
Notify, control terminal parsing sending out notice, by the picture in agreement request server to obtain demographics result.
Including procedure below further, in step s 40,：
S41:By Zheng Tu areas video；
S42:Interaction window travels through and splits each frame；
S43：The cognitron that the image input training that segmentation is completed is completed is judged；
S44：Use rectangle frame standard people's head region；
S45：Rectangle frame is repeated to merge.
Further, in step S60, including procedure below：
S61:Video acquisition；
S62:Reset；
S63：ROI areas are set；
S64：Mark ROI areas；
S65：To ROI region processing；
S66：Cancel ROI.
The additional aspect and advantage of the present invention will be set forth in part in the description, and will partly become from the following description
Obtain substantially, or recognized by the practice of the present invention.
Brief description of the drawings
The above-mentioned and/or additional aspect and advantage of the present invention will become from description of the accompanying drawings below to embodiment is combined
Substantially and be readily appreciated that, wherein：
Fig. 1 shows the flow chart according to the inventive method；
The sub-process figure of the real-time head detection process of Fig. 2 the inventive method；
Fig. 3 is the flow chart of the setting ROI region of the inventive method.
Embodiment
It is below in conjunction with the accompanying drawings and specific real in order to be more clearly understood that the above objects, features and advantages of the present invention
Mode is applied the present invention is further described in detail.It should be noted that in the case where not conflicting, the implementation of the application
Feature in example and embodiment can be mutually combined.
Many details are elaborated in the following description in order to fully understand the present invention, and still, the present invention may be used also
Implemented with being different from other modes described here using other, therefore, protection scope of the present invention is not by described below
Specific embodiment limitation.
With reference to Fig. 1, to an a kind of specific embodiment of demographic method based on video analysis of the present invention
It is specifically described.
The invention provides a kind of demographic method based on video analysis, the video is clapped using camera
At least one set of pedestrian monitoring Sample video taken the photograph and one group of pedestrian's monitor and detection video, as shown in figure 1, the demographic method
Comprise the following steps：
S10：The number of people image in pedestrian's monitoring Sample video is gathered, so that positive sample training image collection is constituted, with described
Those in pedestrian's monitoring Sample video in addition to number of people image are not pedestrian heads but the image conduct of doubtful pedestrian head
Negative sample training image collection, and the positive sample training image collection and the negative sample training image collection are normalized
Into the training input picture of an equal amount of size；The monitoring of statistical number of person the need for gathering in pedestrian's monitor and detection video
Area image frame.Due to the head sample-size disunity of Manual interception, should not as training input picture, we must be by
The picture of interception zooms to an equal amount of size, is normalized as 32（Pixel）×32（Pixel）.
S20：The gradient orientation histogram feature of training input picture and monitor area picture frame is extracted respectively（HOG）；Ladder
Spend direction histogram feature（HOG）The gradient direction of mainly localized region is calculated, and is then described with histogram, i.e.,
A kind of feature descriptor of image local overlapping region in fact.
S30：The off-line training of SVM heads grader is carried out to the gradient orientation histogram feature for training input picture,
A large amount of positive sample training image collections after step S20 processing and the negative sample training image collection are input to supporting vector
It is trained in machine model, obtains the optimum classifier being directed under these training samples.SVMs（SVM）It is a kind of effort
Seek a kind of machine learning method of structural risk minimization；
S40：Real-time head detection, the monitor area picture frame of the detection video sequence after step S20 processing is input to
In the grader trained, by setting detection window moving step length and scaling the dimension scale of abstract image window, detection is allowed
Window is scanned on different positions, differentiates that these windows are that head zone is also non-head according to the result of decision of grader
Portion region, finally marks the position on head in the video sequence.
S50：Target following is carried out using optical flow method, realizes and lasting inspection is carried out to the head movement in video frame image
Survey, matching and the tracking of location updating message.Optical flow method is used for the principle of target following：1）To a continuous sequence of frames of video
Carry out processing 2）For each video sequence, using certain object detection method, the foreground target 3 being likely to occur is detected）
If a certain frame occurs in that foreground target, its representative key point is found.
S60：ROI is set, area-of-interest, detachment's area-of-interest are realized using the setting of the ROI inside opencv
Handled；Area-of-interest can arbitrarily be set according to actual requirement, so that improving system obtains practicality.In order to pass in and out people
The statistics of mouth, detects the number of people, and train the picture process of acquisition under off-line state by the way of vertical camera
Grader, then design and Implement passenger number statistical system.
S70：Statistical number of person, according to the movement locus on head, contrast imports and exports mark and can determine whether, into outgoing direction, to work as moving object
Body, which first encounters import mark and then proceedes to motion, to be encountered exit marking to enter behavior, and when moving object, first to encounter exit marking right
Follow-up reforwarding is dynamic to encounter import labeled as behavior of going out, and finally, judges according to mark and head movement track is imported and exported into trip
For to realize the counting statistics of number.
S80:The picture detected, is saved as img.jpg formatted files by remote inquiry, to pathon journeys while preservation
Sequence sends out a notice, and picture file is saved in into server, then sends notice to remote control terminal, and control terminal parsing is pushed
Notify, by the picture in agreement request server to obtain demographics result.Specifically picture is stored in into Chinese shifting
Dynamic server（OneNet）, then to mobile phone terminal transmission notice, mobile phone terminal parsing sending out notice passes through Http agreements, request
OneNet picture.
This method is counted using computer vision technique to the flow of the people of specific region, it is possible to checked in mobile phone terminal
Not picture in the same time.For the method for human testing, it can significantly avoid detecting asking of blocking mutually of target
Topic.Pedestrian head detection is used based on gradient orientation histogram（HOG）Feature extraction, linear SVM（SVM）Make
For the detection method of grader, pedestrian is tracked using optical flow method.Build on wide variety of video monitoring equipment, into
This low, wide application, statistical accuracy are high.
Further, as shown in Fig. 2 in step s 40, including procedure below：
S41:By Zheng Tu areas video；
S42:Interaction window travels through and splits each frame；
S43：The cognitron that the image input training that segmentation is completed is completed is judged；
S44：Use rectangle frame standard people's head region；
S45：Rectangle frame is repeated to merge.
Further, as shown in figure 3, in step S60, including procedure below：
S61:Video acquisition；
S62:Reset；
S63：ROI areas are set；
S64：Mark ROI areas；
S65：To ROI region processing；
S66：Cancel ROI.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.Within the spirit and principles of the invention, that is made any repaiies
Change, equivalent, improvement etc., should be included within the scope of the present invention.
Claims (4)
1. a kind of people flow rate statistical method based on video analysis, the video is shot using camera from top area
At least one set of pedestrian's monitoring Sample video and one group of pedestrian's monitor and detection video, the demographic method comprise the following steps：
S10：The number of people image in pedestrian's monitoring Sample video is gathered, so that positive sample training image collection is constituted, with described
Those in pedestrian's monitoring Sample video in addition to number of people image are not pedestrian heads but the image conduct of doubtful pedestrian head
Negative sample training image collection, and the positive sample training image collection and the negative sample training image collection are normalized
Into the training input picture of an equal amount of size；The monitoring of statistical number of person the need for gathering in pedestrian's monitor and detection video
Area image frame；
S20：The gradient orientation histogram feature of training input picture and monitor area picture frame is extracted respectively（HOG）；Gradient side
To histogram feature（HOG）The gradient direction of mainly localized region is calculated, and is then described with histogram, i.e., in fact
A kind of feature descriptor of image local overlapping region；
S30：The off-line training of SVM heads grader is carried out to the gradient orientation histogram feature for training input picture, a large amount of
Positive sample training image collection and the negative sample training image collection after step S20 processing are input to SVMs mould
It is trained in type, obtains the optimum classifier being directed under these training samples；SVMs（SVM）It is that a kind of effort is sought
A kind of machine learning method of structural risk minimization；
S40：Real-time head detection, the monitor area picture frame of the detection video sequence after step S20 processing is input to
In the grader trained, by setting detection window moving step length and scaling the dimension scale of abstract image window, detection is allowed
Window is scanned on different positions, differentiates that these windows are that head zone is also non-head according to the result of decision of grader
Portion region, finally marks the position on head in the video sequence；
S50：Target following is carried out using optical flow method, realize the head movement in video frame image is carried out lasting detection,
With the tracking with location updating message；
S60：ROI is set, area-of-interest is realized using the setting of the ROI inside opencv, detachment's area-of-interest is carried out
Processing；
S70：Statistical number of person, according to the movement locus on head, contrast is imported and exported mark and can determine whether into outgoing direction, when moving object is first
Encounter import mark and then proceed to motion and encounter exit marking to enter behavior, when moving object first encounter exit marking then after
Reforwarding is dynamic to encounter import labeled as behavior of going out, and finally, turnover behavior is judged according to mark and head movement track is imported and exported, with
Realize the counting statistics of number.
2. the people flow rate statistical method according to claim 1 based on video analysis, it is characterised in that also including step
S80:The picture detected, is saved as img.jpg formatted files by remote inquiry, is sent out while preservation to pathon programs
One notice, server is saved in by picture file, then sends notice to remote control terminal, and control terminal parsing pushes logical
Know, by the picture in agreement request server to obtain demographics result.
3. the people flow rate statistical method according to claim 1 based on video analysis, it is characterised in that in step s 40,
Including procedure below：
S41:By Zheng Tu areas video；
S42:Interaction window travels through and splits each frame；
S43：The cognitron that the image input training that segmentation is completed is completed is judged；
S44：Use rectangle frame standard people's head region；
S45：Rectangle frame is repeated to merge.
4. the people flow rate statistical method according to claim 3 based on video analysis, it is characterised in that in step S60,
Including procedure below：
S61:Video acquisition；
S62:Reset；
S63：ROI areas are set；
S64：Mark ROI areas；
S65：To ROI region processing；
S66：Cancel ROI.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710225976.2A CN106951885A (en) | 2017-04-08 | 2017-04-08 | A kind of people flow rate statistical method based on video analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710225976.2A CN106951885A (en) | 2017-04-08 | 2017-04-08 | A kind of people flow rate statistical method based on video analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106951885A true CN106951885A (en) | 2017-07-14 |
Family
ID=59475515
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710225976.2A Pending CN106951885A (en) | 2017-04-08 | 2017-04-08 | A kind of people flow rate statistical method based on video analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106951885A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107396047A (en) * | 2017-07-25 | 2017-11-24 | 国网江苏省电力公司南通供电公司 | Inlet/outlet programming count long-distance management system |
| CN107563285A (en) * | 2017-07-27 | 2018-01-09 | 广东数相智能科技有限公司 | The method and device that a kind of speed based on image recognition calculates |
| CN108257420A (en) * | 2018-02-11 | 2018-07-06 | 江苏金海星导航科技有限公司 | Meter people based on camera counts vehicle method, apparatus and system |
| CN108256462A (en) * | 2018-01-12 | 2018-07-06 | 北京航空航天大学 | A kind of demographic method in market monitor video |
| CN108388883A (en) * | 2018-03-16 | 2018-08-10 | 广西师范大学 | A kind of video demographic method based on HOG+SVM |
| CN108615020A (en) * | 2018-04-28 | 2018-10-02 | 东莞市华睿电子科技有限公司 | A kind of floating population number statistical method in video monitoring regional |
| CN108647615A (en) * | 2018-04-28 | 2018-10-12 | 东莞市华睿电子科技有限公司 | A kind of cinema steals a ride the implementation method of early warning |
| CN109101929A (en) * | 2018-08-16 | 2018-12-28 | 新智数字科技有限公司 | A kind of pedestrian counting method and device |
| CN109272487A (en) * | 2018-08-16 | 2019-01-25 | 北京此时此地信息科技有限公司 | The quantity statistics method of crowd in a kind of public domain based on video |
| CN109389016A (en) * | 2017-08-10 | 2019-02-26 | 株式会社日立制作所 | A kind of method and system that the number of people counts |
| CN110599781A (en) * | 2019-09-23 | 2019-12-20 | 潍坊畅讯交通设施工程有限公司 | Traffic flow statistics and recognition system for intelligent city |
| CN110781735A (en) * | 2019-09-18 | 2020-02-11 | 重庆特斯联智慧科技股份有限公司 | Alarm method and system for identifying on-duty state of personnel |
| CN110929626A (en) * | 2019-11-18 | 2020-03-27 | 上海欧冶物流股份有限公司 | Method for determining number of inventory goods, warehouse checking method, device and equipment |
| CN111209781A (en) * | 2018-11-22 | 2020-05-29 | 珠海格力电器股份有限公司 | Method and device for counting number of people in room |
| CN112307936A (en) * | 2020-10-28 | 2021-02-02 | 江苏云从曦和人工智能有限公司 | Passenger flow volume analysis method, system and device based on head and shoulder detection |
| CN113344879A (en) * | 2021-06-09 | 2021-09-03 | 成都之维安科技股份有限公司 | Image target segmentation and color anomaly detection method based on pollution source discharge port |
| CN113392726A (en) * | 2021-05-26 | 2021-09-14 | 上海飞乐音响股份有限公司 | Method, system, terminal and medium for identifying and detecting human head in outdoor monitoring scene |
| CN114022143A (en) * | 2022-01-06 | 2022-02-08 | 深圳爱莫科技有限公司 | Code scanning payment behavior model training method, detection method and processing equipment |
| CN114187666A (en) * | 2021-12-23 | 2022-03-15 | 中海油信息科技有限公司 | Identification method and system for watching mobile phone while walking |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101464946A (en) * | 2009-01-08 | 2009-06-24 | 上海交通大学 | Detection method based on head identification and tracking characteristics |
| CN102799935A (en) * | 2012-06-21 | 2012-11-28 | 武汉烽火众智数字技术有限责任公司 | Human flow counting method based on video analysis technology |
| CN102930287A (en) * | 2012-09-26 | 2013-02-13 | 上海理工大学 | Overlook-based detection and counting system and method for pedestrians |
-
2017
- 2017-04-08 CN CN201710225976.2A patent/CN106951885A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101464946A (en) * | 2009-01-08 | 2009-06-24 | 上海交通大学 | Detection method based on head identification and tracking characteristics |
| CN102799935A (en) * | 2012-06-21 | 2012-11-28 | 武汉烽火众智数字技术有限责任公司 | Human flow counting method based on video analysis technology |
| CN102930287A (en) * | 2012-09-26 | 2013-02-13 | 上海理工大学 | Overlook-based detection and counting system and method for pedestrians |
Non-Patent Citations (1)
| Title |
|---|
| 贾世杰等: "基于人头目标检测的商场人流量自动统计", 《大连交通大学学报》 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107396047A (en) * | 2017-07-25 | 2017-11-24 | 国网江苏省电力公司南通供电公司 | Inlet/outlet programming count long-distance management system |
| CN107563285A (en) * | 2017-07-27 | 2018-01-09 | 广东数相智能科技有限公司 | The method and device that a kind of speed based on image recognition calculates |
| CN109389016A (en) * | 2017-08-10 | 2019-02-26 | 株式会社日立制作所 | A kind of method and system that the number of people counts |
| CN109389016B (en) * | 2017-08-10 | 2023-04-07 | 株式会社日立制作所 | Method and system for counting human heads |
| CN108256462A (en) * | 2018-01-12 | 2018-07-06 | 北京航空航天大学 | A kind of demographic method in market monitor video |
| CN108257420A (en) * | 2018-02-11 | 2018-07-06 | 江苏金海星导航科技有限公司 | Meter people based on camera counts vehicle method, apparatus and system |
| CN108388883A (en) * | 2018-03-16 | 2018-08-10 | 广西师范大学 | A kind of video demographic method based on HOG+SVM |
| CN108615020A (en) * | 2018-04-28 | 2018-10-02 | 东莞市华睿电子科技有限公司 | A kind of floating population number statistical method in video monitoring regional |
| CN108647615A (en) * | 2018-04-28 | 2018-10-12 | 东莞市华睿电子科技有限公司 | A kind of cinema steals a ride the implementation method of early warning |
| CN109101929A (en) * | 2018-08-16 | 2018-12-28 | 新智数字科技有限公司 | A kind of pedestrian counting method and device |
| CN109272487A (en) * | 2018-08-16 | 2019-01-25 | 北京此时此地信息科技有限公司 | The quantity statistics method of crowd in a kind of public domain based on video |
| CN111209781B (en) * | 2018-11-22 | 2023-05-23 | 珠海格力电器股份有限公司 | Method and device for counting indoor people |
| CN111209781A (en) * | 2018-11-22 | 2020-05-29 | 珠海格力电器股份有限公司 | Method and device for counting number of people in room |
| CN110781735A (en) * | 2019-09-18 | 2020-02-11 | 重庆特斯联智慧科技股份有限公司 | Alarm method and system for identifying on-duty state of personnel |
| CN110599781A (en) * | 2019-09-23 | 2019-12-20 | 潍坊畅讯交通设施工程有限公司 | Traffic flow statistics and recognition system for intelligent city |
| CN110929626A (en) * | 2019-11-18 | 2020-03-27 | 上海欧冶物流股份有限公司 | Method for determining number of inventory goods, warehouse checking method, device and equipment |
| CN112307936A (en) * | 2020-10-28 | 2021-02-02 | 江苏云从曦和人工智能有限公司 | Passenger flow volume analysis method, system and device based on head and shoulder detection |
| CN113392726A (en) * | 2021-05-26 | 2021-09-14 | 上海飞乐音响股份有限公司 | Method, system, terminal and medium for identifying and detecting human head in outdoor monitoring scene |
| CN113344879A (en) * | 2021-06-09 | 2021-09-03 | 成都之维安科技股份有限公司 | Image target segmentation and color anomaly detection method based on pollution source discharge port |
| CN114187666A (en) * | 2021-12-23 | 2022-03-15 | 中海油信息科技有限公司 | Identification method and system for watching mobile phone while walking |
| CN114187666B (en) * | 2021-12-23 | 2022-09-02 | 中海油信息科技有限公司 | Identification method and system for watching mobile phone while walking |
| CN114022143A (en) * | 2022-01-06 | 2022-02-08 | 深圳爱莫科技有限公司 | Code scanning payment behavior model training method, detection method and processing equipment |
| CN114022143B (en) * | 2022-01-06 | 2022-04-05 | 深圳爱莫科技有限公司 | Code scanning payment behavior detection method and processing equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106951885A (en) | A kind of people flow rate statistical method based on video analysis | |
| CN109644255B (en) | Method and apparatus for annotating a video stream comprising a set of frames | |
| CN102799935B (en) | Human flow counting method based on video analysis technology | |
| CN110781836A (en) | Human body recognition method and device, computer equipment and storage medium | |
| US20070098222A1 (en) | Scene analysis | |
| CN103824070A (en) | Rapid pedestrian detection method based on computer vision | |
| CN110633612A (en) | Monitoring method and system for inspection robot | |
| CN111753601B (en) | Image processing method, device and storage medium | |
| CN111325048A (en) | Personnel gathering detection method and device | |
| Saon et al. | Cloud-based people counter | |
| CN111008574A (en) | Key person track analysis method based on body shape recognition technology | |
| CN111476160A (en) | Loss function optimization method, model training method, target detection method, and medium | |
| CN112465854A (en) | Unmanned aerial vehicle tracking method based on anchor-free detection algorithm | |
| CN112084892B (en) | Road abnormal event detection management device and method thereof | |
| CN109903308B (en) | Method and device for acquiring information | |
| CN116311166A (en) | Traffic obstacle recognition method and device and electronic equipment | |
| CN113537019A (en) | Detection method for identifying wearing of safety helmet of transformer substation personnel based on key points | |
| CN112488042A (en) | Pedestrian traffic bottleneck discrimination method and system based on video analysis | |
| CN110580708B (en) | Rapid movement detection method and device and electronic equipment | |
| Dinh et al. | Development of a tracking-based system for automated traffic data collection for roundabouts | |
| CN105809183A (en) | Video-based human head tracking method and device thereof | |
| CN114067396A (en) | Vision learning-based digital management system and method for live-in project field test | |
| CN114092956A (en) | Store passenger flow statistical method and device, computer equipment and storage medium | |
| CN113963318B (en) | People flow statistical method and device, electronic equipment and storage medium | |
| CN111860100B (en) | Pedestrian number determining method and device, electronic equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170714 |