CN108416254A - A kind of statistical system and method for stream of people's Activity recognition and demographics - Google Patents
A kind of statistical system and method for stream of people's Activity recognition and demographics Download PDFInfo
- Publication number
- CN108416254A CN108416254A CN201810045934.5A CN201810045934A CN108416254A CN 108416254 A CN108416254 A CN 108416254A CN 201810045934 A CN201810045934 A CN 201810045934A CN 108416254 A CN108416254 A CN 108416254A
- Authority
- CN
- China
- Prior art keywords
- real
- infrared
- demographics
- information
- people
- 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
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of statistical systems and method for stream of people's Activity recognition and demographics,The system includes data acquisition device,Message handler and presentation of information interactive device,Data acquisition device is worked by the image preprocessing that probe unit acquires real time monitoring information and does bottom,The image of high quality is provided for message handler,Message handler realizes the segmentation of crowd in the form of module encapsulation,Feature extraction,Cluster,The adaptive training of demographics model and last real-time assessment,The information such as real-time scene demographics are provided for display interactive device,Show that interactive device shows scene information and real-time demographic information by display console,The alarm threshold value set in advance according to user does corresponding actions,And the interactive interface being adjusted to systematic parameter is provided a user in the process of running.The present invention has higher real-time, robustness and accuracy, is suitable for region security system.
Description
Technical field
The present invention relates to high density stream of people situation and statistical methods, and in particular to one kind being used for stream of people's Activity recognition and number
The statistical system and method for statistics belong to computer vision, image procossing, artificial intelligence Video Analysis Technology field, enhancing
The ability of target in complex environment detection, is suitable for region security system, especially suitable for the row such as invading, hovering in the stream of people
For identification, the behavior to paying close attention to personage is detected, tracks and is prevented.
Background technology
Stream of people's statistics and stream of people's status analysis (crowd counts and crowd state analysis) under high density case have its heavy
The social effect and application value wanted.Utilize the stream of people's statistics and stream of people's condition information in region：Administrative staff can rationally adjust
Man power and material is spent, the configuration of resource is optimized；Good forewarning function can be played to a series of social security problems.In view of crowd
For demographics in the significance of intelligent video monitoring and crowd behaviour analysis field, which causes state, inside and outside numerous aspects
Research, propose a variety of methods.In recent years, the famous periodical of computer vision field has a large amount of related Dense crowds to count
And the relevant article of crowd behaviour intellectual analysis based on this and discussion.Especially international advanced video and recording monitor
Forum AVSS can have the special meeting specifically for population analysis every year.In China, to crowd based on computer vision point
Phase separation closes special topic and is also discussed more.Preferable effect is obtained, and is applied in important area security system.But due to complexity
The separation problem of background and target is difficult to solve, and algorithm is more complex, and rate of false alarm is higher, the applied field of people stream counting and statistical item
Scape is confined to the inlet and outlet in region substantially.
Invention content
For the defects in the prior art, the object of the present invention is to provide one kind being used for stream of people's Activity recognition and demographics
Statistical system and method.
The present invention is realized according to following technical scheme：
A kind of statistical system for stream of people's Activity recognition and demographics, the system are based on human body head feature and red
Outer feature, which is characterized in that including data acquisition device, message handler and presentation of information interactive device, the data
Harvester acquires real-time monitoring images information by probe unit and completes the pretreatment of image by image pre-processing unit,
Pretreated image is provided for message handler to upper layer, described information processor realizes crowd in the form of module encapsulation
Segmentation, feature extraction, cluster, the adaptive training of demographics model and last real-time assessment, upwards for display hand over
Mutual device provides real-time scene demographic information, the display interactive device by display console display scene information with
And real-time demographic information, while the alarm threshold value set in advance according to user does corresponding actions, and running
The interactive interface being adjusted to systematic parameter is provided a user in the process.
In above-mentioned technical proposal, the probe unit includes video surveillance devices and infrared electro monitoring device, described red
Outer photoelectric monitor equipment uses sensitive wave length for 8-14 μm of infrared detector, and the video surveillance devices, which are selected, has dim light high
The equipment detected clearly is day and night detected with described with the cooperation of infrared electro monitor.
In above-mentioned technical proposal, described information processor include crowd characteristic processing module, self-adaption cluster module, in real time
Prediction module.
A kind of statistical method for stream of people's Activity recognition and demographics utilizes claim 1-3 any one of them
What statistical system was realized, which is characterized in that include the following steps：
Step S1：Human body target detection in infrared, visible light video image based on human body infrared characteristic, using highlighted
The method of pixel detection, edge detection and parallax triplicity obtains target area；
Step S2：It is imaged using infrared heat, is analyzed by the intensity profile to infrared hybrid optical system；
Step S3：There are Spline smoothing or the pixel extraction of shape variation to come out gray value in image, then by the company of refinement
Continuous edge pixel point is assembled into a profile, carries out object edge extraction；
Step S4：The parameterized template of the number of people is established according to the feature on head, and target is matched using the template；
Step S5：Multiple features headform is established according to the profile of target and parallax information.
Compared with prior art, the present invention has following advantageous effect：
There is the present invention higher real-time, robustness and accuracy, method to enhance target in complex environment detection
Ability is suitable for region security system, especially suitable for the Activity recognitions such as invading, hovering in the stream of people, to paying close attention to the row of personage
To be detected, being tracked and prevented.
Description of the drawings
Upon reading the detailed description of non-limiting embodiments with reference to the following drawings, other feature of the invention,
Objects and advantages will become more apparent upon：
Fig. 1 is the system architecture diagram of the present invention.
Specific implementation mode
With reference to specific embodiment, the present invention is described in detail.Following embodiment will be helpful to the technology of this field
Personnel further understand the present invention, but the invention is not limited in any way.It should be pointed out that the ordinary skill of this field
For personnel, without departing from the inventive concept of the premise, several changes and improvements can also be made.These belong to the present invention
Protection domain.
Fig. 1 is the system architecture diagram of the present invention, as shown in Figure 1, a kind of system for stream of people's Activity recognition and demographics
Meter systems, the system are based on human body head feature and infrared signature, which is characterized in that including data acquisition device, information
Processor and presentation of information interactive device, the data acquisition device acquire real-time monitoring images information simultaneously by probe unit
The pretreatment that image is completed by image pre-processing unit, pretreated image is provided to upper layer for message handler, described
Message handler realized in the form of module encapsulation the segmentation of crowd, feature extraction, cluster, demographics model it is adaptive
Training and last real-time assessment provide real-time scene demographic information, the display for display interactive device upwards
Interactive device shows scene information and real-time demographic information by display console, while being arranged in advance according to user
Good alarm threshold value does corresponding actions, and provides a user connect to the interaction that systematic parameter is adjusted in the process of running
Mouthful.
In a specific embodiment of the present invention, the probe unit includes that video surveillance devices and infrared electro monitoring are set
Standby, the infrared electro monitoring device uses sensitive wave length for 8-14 μm of infrared detector, the video surveillance devices selection
Equipment with the detection of dim light high definition is day and night detected with described with the cooperation of infrared electro monitor.
Described information processor includes crowd characteristic processing module, self-adaption cluster module, real-time prediction module.And these
Module is it is known in the art, being just not described in detail herein.
A kind of statistical method of high density stream of people's situation based on human body head feature and infrared signature proposed by the present invention,
It is realized, is included the following steps using above-mentioned statistical system：
A kind of statistical method for stream of people's Activity recognition and demographics utilizes claim 1-3 any one of them
What statistical system was realized, which is characterized in that include the following steps：
Step S1：Human body target detection in infrared, visible light video image based on human body infrared characteristic, using highlighted
The method of pixel detection, edge detection and parallax triplicity obtains target area；
Step S2：It is imaged using infrared heat, is analyzed by the intensity profile to infrared hybrid optical system；
Step S3：There are Spline smoothing or the pixel extraction of shape variation to come out gray value in image, then by the company of refinement
Continuous edge pixel point is assembled into a profile, carries out object edge extraction；
Step S4：The parameterized template of the number of people is established according to the feature on head, and target is matched using the template, is reached
To quickly in real time；
Step S5：Multiple features headform is established according to the profile of target and parallax information, is reduced using single features
Matching error caused by model has higher real-time, robustness and accuracy.
Embodiment 1
Human body target detection method in infrared, visible light video image based on human body infrared characteristic.
The method of visible images is difficult to solve the separation problem of complex background and target, and Infrared video image utilizes heat
Imaging compares visible light video image, is not easily susceptible to the influence of light, therefore, the background in Infrared video image is fairly simple.
Infra-red radiation is related with the transmission of thermal energy, and human body abides by infrared spoke as a radiation source as other objects
Penetrate law.But human body has the essential characteristic of vital movement as a lived high homoiotherm body.Human body infrared
Radiation and energetic supersession, body heat balance, body heat regulation and the institutional framework of body have close inner link, have it specific
Physiological mechanism and architecture basics, radiant power and the absolute temperature 310 of normal humanoThe black matrix of K is similar.No matter the colour of skin, compare spoke
The rate of penetrating is about 0.99, illustrates that human body has very high radiant power.The peak emission wavelength of black matrix is with black matrix absolute temperature at anti-
Than (λ m indicate the peak wavelength of black matrix to i.e. λ mT=C, and T is the absolute temperature of black matrix, and c is a constant, are equal to 2897.8 μ
m.K).That is, the heat radiation of object is wavelength and the function of temperature, temperature is higher, and radiation energy is bigger, and peak wavelength is then
It is moved to shortwave direction.According to this law, only it is to be understood that the absolute temperature of object, can calculate its peak value radiated wave
Long, normal human is 37 DEG C, that is, 310oK (K indicates absolute temperature scale, it and poor 273.15 degree Celsius), is calculated by formula, people
Body peak emission wavelength λ m=9.348 μm.
Since human eye can only be sensitive to 0.4~0.7 μm of visible light wave range, and the infrared detector of various different materials
(thermosensitive type, photon type), can detect 0.7-14 μm of infrared band.The present invention is limited to existing according to human body peak emission wavelength
There is infrared detector finished product, the infrared detector that sensitive wave length is 8-14 μm is selected to detect human infrared radiation.By a series of
Signal processing, thermal radiation signal is converted into thermal-induced imagery that is visual and can quantifying, with video surveillance devices can
Light-exposed video image fusion.
Embodiment 2
High density stream of people situation based on head feature and statistical method；
Since the attitudes vibration of human body is larger, need to establish using the matched stream of people's situation of shape and statistical method a large amount of
Shape template, it is also contemplated that how to extract the integrity profile of target.But it can be seen that by analyzing human external feature：Human body head
Portion's feature is more apparent, the feature combined with shoulder using head, and number of people top shape and feature are relatively fixed, and establishes a people
Body head feature model is positioned target using the model in the image of acquisition, therefore is identified using head feature, can reach fast
It is fast real-time.
Embodiment 3
Potential target region segmentation in infrared image is come out, is accurately positioned.
It is to come out the potential target region segmentation in infrared image to draw a circle to approve this region segmentation, in order to subsequent target
Detection operation.ROI segmentation steps include high luminance pixels detection, morphology processing, label connected region and region merging technique 4
Sub-steps.Wherein connected region and region merging technique is marked to be completed by integrated treatment unit related algorithm.
The invention mainly comprises high luminance pixels detection and morphology processings：
1. high luminance pixels detect
In order to be effectively split to drawing a circle to approve this region ROI, the intensity profile of infrared hybrid optical system is carried out first
Research.Typically, since infrared image is insensitive to light, only to heat sensitive, and the environment temperature of a scene is ratio
More uniform, so comparing visible images, the rule of intensity profile is more apparent.
2. morphology processing
Object edge is extracted, and the edge of image is the part that grey scale change is violent in image, is analysis image shape feature
Basis.Edge extracting is exactly to have Spline smoothing or the pixel of shape variation to carry gray value in image by certain mathematical method
It takes out.Ideal Image Edge-Detection the result is that refinement continuous boundary pixel, then by these continuous pixel groups
It dresses up as a profile, i.e. contours extract.The effect of Image Edge-Detection directly affects subsequent edge analysis.Edge of the present invention
Detection algorithm utilizes the derivative calculations of Gaussian filter, and improves randomized hough transform to determine profile parameters.
It is exactly in next step target detection after completing delineation this region R01 segmentations.Shape matching is most straightforward approach,
The parameterized template of the number of people is established according to the feature on head, and target is matched using the template；
Target area is obtained using the method for high luminance pixels detection, edge detection and parallax triplicity, is then passed through
Head matching carries out target detection；It is then comprehensive using histogram, target similitude using the Target Segmentation method based on projection
And the sorting technique that is combined of edge feature the methods of is detected target.
Pedestrian counting problems faced, which is carried out, using number of people target is primarily present in the real-time of algorithm, robustness and accurate
Property 3 aspects, single detection means is difficult to meet actual requirement, needs to merge a variety of methods to improve the performance needles of system
To the deficiency that Detection accuracy in template matching method is easily affected by the external environment, pass through foundation to number of people parameter model and mesh
The research of 2 aspects of track algorithm is marked, the present invention establishes a kind of pedestrian and static crowd's number system, is proposed for this system
Algorithm inherit the good advantage of template matching method real-time, Average Accuracy up to 95% or more in terms of target identification, this
Invention establishes multiple features headform according to the profile and parallax information of target, reduces using caused by single features model
With error.
Embodiment 4
Real-time number counting and crowd state analysis, Dense crowd meter are reached using the method that detection, tracking are combined
The relevant informations such as number and pedestrian tracking.Crowd in scene is reached using the method for pedestrian tracking track and the behaviors such as is detained, hovers
Analysis.
Track algorithm is initialized with the target area that the number of people detects, is pre-processed to tracing area
Afterwards, using improved Mean Shift (mean shift vectors) track algorithms to target into line trace.There is pedestrian in scene
It is interior be detained, hover when, system can effectively be judged, to eliminate the error of repeat count, improve statistics
Accuracy rate.And according to making analysis situations such as requirement to being detained, hovering.
There is the present invention higher real-time, robustness and accuracy, method to enhance target in complex environment detection
Ability is suitable for region security system, especially suitable for the Activity recognitions such as invading, hovering in the stream of people, to paying close attention to the row of personage
To be detected, being tracked and prevented.
One skilled in the art will appreciate that in addition to realizing system provided by the invention in a manner of pure computer readable program code
It, completely can be by the way that method and step be carried out programming in logic come so that the present invention provides and its other than each device, module, unit
System and its each device, module, unit with logic gate, switch, application-specific integrated circuit, programmable logic controller (PLC) and embedding
Enter the form of the controller that declines etc. to realize identical program.So system provided by the invention and its every device, module, list
Member is considered a kind of hardware component, and also may be used for realizing the device of various programs, module, unit to include in it
To be considered as the structure in hardware component；It can also will be considered as realizing the device of various functions, module, unit either real
The software program of existing method can be the structure in hardware component again.
Specific embodiments of the present invention are described above.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make a variety of changes or change within the scope of the claims, this not shadow
Ring the substantive content of the present invention.In the absence of conflict, the feature in embodiments herein and embodiment can arbitrary phase
Mutually combination.
Claims (4)
1. a kind of statistical system for stream of people's Activity recognition and demographics, which is based on human body head feature and infrared
Feature, which is characterized in that including data acquisition device, message handler and presentation of information interactive device, the data are adopted
Acquisition means acquire real-time monitoring images information by probe unit and complete the pretreatment of image by image pre-processing unit, to
Upper layer provides pretreated image for message handler, and described information processor realizes crowd's in the form of module encapsulation
Segmentation, feature extraction, cluster, the adaptive training of demographics model and last real-time assessment are upwards display interaction
Device provides real-time scene demographic information, the display interactive device by display console display scene information and
Real-time demographic information, while the alarm threshold value set in advance according to user does corresponding actions, and running
The interactive interface being adjusted to systematic parameter is provided a user in journey.
2. the statistical system of a kind of stream of people's Activity recognition and demographics according to claim 1, which is characterized in that described
Probe unit includes video surveillance devices and infrared electro monitoring device, the infrared electro monitoring device use sensitive wave length for
8-14 μm of infrared detector, the video surveillance devices select the equipment with the detection of dim light high definition, with described and infrared light
The cooperation of pyroelectric monitor device is day and night detected.
3. the statistical system of a kind of stream of people's Activity recognition and demographics according to claim 1, which is characterized in that described
Message handler includes crowd characteristic processing module, self-adaption cluster module, real-time prediction module.
4. a kind of statistical method for stream of people's Activity recognition and demographics is united using claim 1-3 any one of them
What meter systems were realized, which is characterized in that include the following steps：
Step S1：Human body target detection in infrared, visible light video image based on human body infrared characteristic, using high luminance pixels
The method of detection, edge detection and parallax triplicity obtains target area；
Step S2：It is imaged using infrared heat, is analyzed by the intensity profile to infrared hybrid optical system；
Step S3：There are Spline smoothing or the pixel extraction of shape variation to come out gray value in image, then by the continuous side of refinement
Edge pixel is assembled into a profile, carries out object edge extraction；
Step S4：The parameterized template of the number of people is established according to the feature on head, and target is matched using the template；
Step S5：Multiple features headform is established according to the profile of target and parallax information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810045934.5A CN108416254A (en) | 2018-01-17 | 2018-01-17 | A kind of statistical system and method for stream of people's Activity recognition and demographics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810045934.5A CN108416254A (en) | 2018-01-17 | 2018-01-17 | A kind of statistical system and method for stream of people's Activity recognition and demographics |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108416254A true CN108416254A (en) | 2018-08-17 |
Family
ID=63125991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810045934.5A Pending CN108416254A (en) | 2018-01-17 | 2018-01-17 | A kind of statistical system and method for stream of people's Activity recognition and demographics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108416254A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109819208A (en) * | 2019-01-02 | 2019-05-28 | 江苏警官学院 | A kind of dense population security monitoring management method based on artificial intelligence dynamic monitoring |
| CN110490125A (en) * | 2019-08-15 | 2019-11-22 | 成都睿晓科技有限公司 | A kind of fueling area service quality detection system detected automatically based on gesture |
| CN110675447A (en) * | 2019-08-21 | 2020-01-10 | 电子科技大学 | People counting method based on combination of visible light camera and thermal imager |
| CN111566660A (en) * | 2018-11-28 | 2020-08-21 | 北京嘀嘀无限科技发展有限公司 | System and method for detecting number of persons in vehicle |
| CN112308193A (en) * | 2020-10-29 | 2021-02-02 | 山西大学 | Station ticket checking entrance people flow data collection device |
| CN112422915A (en) * | 2020-11-18 | 2021-02-26 | 珠海格力电器股份有限公司 | Method and device for monitoring number of people, electronic equipment and storage medium |
| WO2022022368A1 (en) * | 2020-07-28 | 2022-02-03 | 宁波环视信息科技有限公司 | Deep-learning-based apparatus and method for monitoring behavioral norms in jail |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000046743A1 (en) * | 1999-02-01 | 2000-08-10 | Beonic Corporation Pty Ltd | Object recognition and tracking system |
| CN101398894A (en) * | 2008-06-17 | 2009-04-01 | 浙江师范大学 | Automobile license plate automatic recognition method and implementing device thereof |
| CN101739550A (en) * | 2009-02-11 | 2010-06-16 | 北京智安邦科技有限公司 | Method and system for detecting moving objects |
| CN102750539A (en) * | 2012-06-06 | 2012-10-24 | 深圳市海亿达能源科技股份有限公司 | Device and method for counting regional space population |
| CN102968646A (en) * | 2012-10-25 | 2013-03-13 | 华中科技大学 | Plate number detecting method based on machine learning |
| CN204667419U (en) * | 2014-12-15 | 2015-09-23 | 成都凌感科技有限公司 | The LCDs of a kind of stream of people's statistics and behavioural analysis |
| CN105787456A (en) * | 2016-03-02 | 2016-07-20 | 江苏大学 | Method for detecting pedestrians in night far infrared images |
| CN106897698A (en) * | 2017-02-24 | 2017-06-27 | 常州常工电子科技股份有限公司 | Classroom number detection method and system based on machine vision Yu binocular coordination technique |
| CN107480585A (en) * | 2017-07-06 | 2017-12-15 | 西安电子科技大学 | Object detection method based on DPM algorithms |
-
2018
- 2018-01-17 CN CN201810045934.5A patent/CN108416254A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000046743A1 (en) * | 1999-02-01 | 2000-08-10 | Beonic Corporation Pty Ltd | Object recognition and tracking system |
| CN101398894A (en) * | 2008-06-17 | 2009-04-01 | 浙江师范大学 | Automobile license plate automatic recognition method and implementing device thereof |
| CN101739550A (en) * | 2009-02-11 | 2010-06-16 | 北京智安邦科技有限公司 | Method and system for detecting moving objects |
| CN102750539A (en) * | 2012-06-06 | 2012-10-24 | 深圳市海亿达能源科技股份有限公司 | Device and method for counting regional space population |
| CN102968646A (en) * | 2012-10-25 | 2013-03-13 | 华中科技大学 | Plate number detecting method based on machine learning |
| CN204667419U (en) * | 2014-12-15 | 2015-09-23 | 成都凌感科技有限公司 | The LCDs of a kind of stream of people's statistics and behavioural analysis |
| CN105787456A (en) * | 2016-03-02 | 2016-07-20 | 江苏大学 | Method for detecting pedestrians in night far infrared images |
| CN106897698A (en) * | 2017-02-24 | 2017-06-27 | 常州常工电子科技股份有限公司 | Classroom number detection method and system based on machine vision Yu binocular coordination technique |
| CN107480585A (en) * | 2017-07-06 | 2017-12-15 | 西安电子科技大学 | Object detection method based on DPM algorithms |
Non-Patent Citations (6)
| Title |
|---|
| 中国铁路总公司: "《高速铁路通信技术》", 30 April 2016 * |
| 全国大学生电子设计竞赛信息安全技术专题邀请赛组委会: "《全国大学生电子设计竞赛信息安全技术专题邀请赛优秀作品选编 第三届》", 31 May 2015, 北京：北京理工大学出版社 * |
| 北方工业大学: "《优秀毕业论文选集 本科2009届》", 30 September 2009, 北方工业大学教务处 * |
| 张勇刚: "《道路交通控制技术及应用》", 31 August 2016, 北京：中国人民公安大学出版社 * |
| 朱诺: "《行人交通安全 基于视频监测和元胞自动机的人群疏散机理研究》", 31 October 2017 * |
| 段大高等: "《图像处理与应用》", 31 January 2013, 北京：北京邮电大学出版社 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111566660A (en) * | 2018-11-28 | 2020-08-21 | 北京嘀嘀无限科技发展有限公司 | System and method for detecting number of persons in vehicle |
| CN109819208A (en) * | 2019-01-02 | 2019-05-28 | 江苏警官学院 | A kind of dense population security monitoring management method based on artificial intelligence dynamic monitoring |
| CN109819208B (en) * | 2019-01-02 | 2021-01-12 | 江苏警官学院 | Intensive population security monitoring management method based on artificial intelligence dynamic monitoring |
| CN110490125A (en) * | 2019-08-15 | 2019-11-22 | 成都睿晓科技有限公司 | A kind of fueling area service quality detection system detected automatically based on gesture |
| CN110490125B (en) * | 2019-08-15 | 2023-04-18 | 成都睿晓科技有限公司 | Oil filling area service quality detection system based on gesture automatic detection |
| CN110675447A (en) * | 2019-08-21 | 2020-01-10 | 电子科技大学 | People counting method based on combination of visible light camera and thermal imager |
| WO2022022368A1 (en) * | 2020-07-28 | 2022-02-03 | 宁波环视信息科技有限公司 | Deep-learning-based apparatus and method for monitoring behavioral norms in jail |
| CN112308193A (en) * | 2020-10-29 | 2021-02-02 | 山西大学 | Station ticket checking entrance people flow data collection device |
| CN112422915A (en) * | 2020-11-18 | 2021-02-26 | 珠海格力电器股份有限公司 | Method and device for monitoring number of people, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108416254A (en) | A kind of statistical system and method for stream of people's Activity recognition and demographics | |
| Radman et al. | Automated segmentation of iris images acquired in an unconstrained environment using HOG-SVM and GrowCut | |
| CN102473238B (en) | Method and system for image analysis | |
| US20080144885A1 (en) | Threat Detection Based on Radiation Contrast | |
| Patruno et al. | People re-identification using skeleton standard posture and color descriptors from RGB-D data | |
| Zhang et al. | Pedestrian detection based on improved LeNet-5 convolutional neural network | |
| US9412025B2 (en) | Systems and methods to classify moving airplanes in airports | |
| De Santis et al. | Robust real time eye tracking for computer interface for disabled people | |
| CN111666920B (en) | Target article wearing detection method and device, storage medium and electronic device | |
| CN104598888B (en) | A kind of recognition methods of face gender | |
| Wang et al. | A novel facial thermal feature extraction method for non-contact healthcare system | |
| Seo et al. | A yolo-based separation of touching-pigs for smart pig farm applications | |
| Shi et al. | Robust image processing algorithm for computational resource limited smart apple sunburn sensing system | |
| Zaidi et al. | Video anomaly detection and classification for human activity recognition | |
| Castillo et al. | Smart environment architecture for robust people detection by infrared and visible video fusion | |
| Cardone et al. | Automated warping procedure for facial thermal imaging based on features identification in the visible domain | |
| Bekhouche et al. | Spatiotemporal CNN with Pyramid Bottleneck Blocks: Application to eye blinking detection | |
| Maksymiv et al. | Deep convolutional network for detecting probable emergency situations | |
| Larregui et al. | An image processing pipeline to segment iris for unconstrained cow identification system | |
| Liang et al. | Real time hand movement trajectory tracking for enhancing dementia screening in ageing deaf signers of British sign language | |
| Qi et al. | Extraction and Classification of Image Features for Fire Recognition Based on Convolutional Neural Network. | |
| Peng et al. | [Retracted] Helmet Wearing Recognition of Construction Workers Using Convolutional Neural Network | |
| Allegra et al. | Exploiting egocentric vision on shopping cart for out-of-stock detection in retail environments | |
| Andrushia et al. | Visual attention-based leukocyte image segmentation using extreme learning machine | |
| Montazeri et al. | Automatically eye detection with different gray intensity image conditions |
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 |