CN109447042A - The system and method for top-type passenger flow monitor processing is realized based on stereovision technique - Google Patents
The system and method for top-type passenger flow monitor processing is realized based on stereovision technique Download PDFInfo
- Publication number
- CN109447042A CN109447042A CN201811542849.6A CN201811542849A CN109447042A CN 109447042 A CN109447042 A CN 109447042A CN 201811542849 A CN201811542849 A CN 201811542849A CN 109447042 A CN109447042 A CN 109447042A
- Authority
- CN
- China
- Prior art keywords
- passenger flow
- flow monitor
- video decoding
- processing
- type passenger
- 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
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000012545 processing Methods 0.000 title claims abstract description 32
- 230000003287 optical effect Effects 0.000 claims abstract description 5
- 238000001514 detection method Methods 0.000 claims description 16
- 238000005259 measurement Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 abstract description 18
- 238000012544 monitoring process Methods 0.000 abstract description 7
- 238000001914 filtration Methods 0.000 abstract description 5
- 238000009434 installation Methods 0.000 abstract description 5
- 238000009877 rendering Methods 0.000 abstract description 3
- 230000008901 benefit Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000011282 treatment Methods 0.000 description 4
- 238000012937 correction Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 239000000047 product Substances 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000007795 chemical reaction product Substances 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005206 flow analysis Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to a kind of systems for realizing the processing of top-type passenger flow monitor based on stereovision technique, including binocular camera group, for acquiring passenger flow data;Video decoding circuit group is connected with the binocular camera group, for the camera data of the binocular camera group acquisition to be carried out video decoding;High-performance processor is connected with the video decoding circuit group, for optical scene information to be converted to digital video image information.The invention further relates to a kind of methods of top-type passenger flow monitor processing realized based on stereovision technique.Using the system and method, the interference of some non-human factors is effectively eliminated using 3D rendering identification technology for real-time passenger flow statistics and monitoring, solves the problems, such as to substantially increase the accuracy of passenger flow statistics and monitoring side by side;The highly filtered technology of personnel is realized, customized the personnel's height for needing filtration counts can be set;The present invention uses contactless statistics, and device installation disguise property is good, does not influence the impression of disengaging personnel.
Description
Technical field
The present invention relates to data collecting fields, more particularly to passenger flow monitor field, in particular to one kind are based on solid in real time
Vision technique realizes the system and method for top-type passenger flow monitor processing.
Background technique
Currently, traditional automatic passenger flow statistics (also referred to as stream of people's statistics or demographics) system can be divided by method of counting division
Are as follows: mechanical railing type, pressure sensing type and infrared detection formula etc. are several, but there are some disadvantages for each.
Mechanical railing type forces pedestrian to pass sequentially through and count using mechanical railing.The installation of its device is complicated, cost
It is high, and cannot freely enter and leave, there is certain invasive to user;
Pressure sensing type passes through the disengaging situation of pressure sensor senses target.Its precision is not high, cannot handle two people or
Situations such as more people enter and leave simultaneously;
Infrared detection technology is more relatively advanced, it obtains target disengaging situation using infrared facility.It is to light
Variation is sensitive, cannot distinguish between the complex situations such as pedestrian and non-pedestrian target.
With the research of machine vision and technical field of image processing and deepening continuously for application work, it is based on machine vision
Passenger flow statistical method have the characteristics that stablize, efficiently, intelligence, hardware resource reusable, thus be gradually widely used.
Using the method for video image processing of mechanical vision inspection technology, passenger flow statistics can be preferably solved the problems, such as, and machine vision
The case where facing is more complicated, has many classification processing methods according to the difference of specific environment and analysis object.
Currently, in the market there are two types of people stream counting mode is general, one is based on monocular-camera, one is be based on
Binocular camera.Wherein monocular-camera mainly detects pedestrian by the head and shoulder model of training of human, and speed is slow, and accuracy rate is not
It is high.And binocular camera passes through binocular parallax range measurement principle, the depth map of available scene.Depth map intuitively reflects field
The distance of object distance camera in scape.And in the application scenarios such as the entrance in bus and market, pedestrian is apparently higher than scene
In other objects, therefore, using binocular camera generate depth map scheme solves the problems, such as pedestrian detection, have using improve examine
The accuracy rate of survey lays good basis for subsequent count tracking.
Summary of the invention
The purpose of the present invention is overcoming the above-mentioned prior art, provide it is a kind of meet highly filtered, precision is high,
The system and method that the processing of top-type passenger flow monitor is realized based on stereovision technique of good concealment.
To achieve the goals above, the system of the invention that the processing of top-type passenger flow monitor is realized based on stereovision technique
And method is as follows:
The system for realizing the processing of top-type passenger flow monitor based on stereovision technique, is mainly characterized by, described is
System includes:
Binocular camera group, for acquiring passenger flow data;
Video decoding circuit group is connected with the binocular camera group, for adopting the binocular camera group
The camera data of collection carry out video decoding;
High-performance processor is connected with the video decoding circuit group, for optical scene information to be converted to number
Word video image information.
Preferably, the binocular camera group includes:
First camera module is connected with the video decoding circuit group, for passing through binocular parallax range measurement principle
Obtain the depth map of scene;
Second camera module is connected with the video decoding circuit group, for passing through binocular parallax range measurement principle
Obtain the depth map of scene.
Preferably, the video decoding circuit group includes:
First video decoding circuit is connected with first camera module, for by first camera
The video data of module acquisition carries out video decoding;
Second video decoding circuit is connected with the second camera module, for by the second camera
The video data of module acquisition carries out video decoding.
Preferably, the system further includes external interface, it is connected with the high-performance processor, for realizing number
According to transmission.
Preferably, the external interface includes Ethernet interface and power interface, with the high-performance processor
It is connected, for realizing data transmission.
Preferably, the system further includes FLASH memory, it is connected with the high-performance processor, for depositing
Store up data.
Preferably, the system further includes DDR memory, it is connected with the high-performance processor, for storing
Data.
Preferably, the system further includes CPLD circuit, it is connected with the high-performance processor.
Preferably, the system further includes RTC circuit, it is connected with the high-performance processor.
This realizes the method for the top-type passenger flow monitor processing based on stereovision technique using above system, main special
Point be, the method the following steps are included:
(1) Image Acquisition is carried out by binocular camera group, calculates step by the way that preset inside and outside parameter and calibration are three-dimensional
The image that year acquires is corrected;
(2) step is calculated by Stereo matching and generates depth map;
(3) pedestrian detection is carried out by the depth map, and according to the relationship between height and depth map brightness to inspection
The personage of survey is modified;
(4) it determines pedestrian's number, calculates relevant attribute.
Preferably, Image Acquisition in the step (1) includes testing the speed stage Image Acquisition and product stage image is adopted
Collection.
Preferably, it includes that BM calculates step that the Stereo matching in the step (2), which calculates step,.
Preferably, the relevant attribute in the step (4) includes average height and position.
Using the system and method for the invention for realizing the processing of top-type passenger flow monitor based on stereovision technique, it is used for
Real-time passenger flow statistics and monitoring are effectively eliminated the interference of some non-human factors, are solved simultaneously using 3D rendering identification technology
Row's problem substantially increases the accuracy of passenger flow statistics and monitoring;The present invention realizes the highly filtered technology of personnel, can make by oneself
Justice setting needs personnel's height of filtration counts;The present invention use contactless statistics, device installation disguise property is good, do not influence into
The impression of personnel out.Expection of the present invention can be public in densely populated places such as subway station, airport, market, museum and core scenic spots
Place uses.Can assist manager efficiently, accurately and timely analyze stream of people's situation, reasonable management and distribute public and manpower
Resource promotes public safety, avoids public accident, with intelligent event store function, provides evidence for legal dispute, generates huge
Economic benefit and social benefit.
Detailed description of the invention
Fig. 1 is the schematic block diagram of the system of the invention that the processing of top-type passenger flow monitor is realized based on stereovision technique.
Fig. 2 is the first of the device of the system of the invention that the processing of top-type passenger flow monitor is realized based on stereovision technique
Kind embodiment structure schematic diagram.
Fig. 3 is the second of the device of the system of the invention that the processing of top-type passenger flow monitor is realized based on stereovision technique
Kind embodiment structure schematic diagram.
Fig. 4 is the stream of people's statistics for the method that realization of the invention is handled based on the top-type passenger flow monitor of stereovision technique
Flow chart of steps.
Fig. 5 is the pedestrian detection for the method that realization of the invention is handled based on the top-type passenger flow monitor of stereovision technique
Algorithm frame figure.
Fig. 6 is the height and depth for the method that realization of the invention is handled based on the top-type passenger flow monitor of stereovision technique
Spend the relational graph of figure gray value.
Specific embodiment
It is further to carry out combined with specific embodiments below in order to more clearly describe technology contents of the invention
Description.
Of the invention system that should realize the processing of top-type passenger flow monitor based on stereovision technique, wherein described is
System includes:
Binocular camera group, for acquiring passenger flow data;
Video decoding circuit group is connected with the binocular camera group, for adopting the binocular camera group
The camera data of collection carry out video decoding;
High-performance processor is connected with the video decoding circuit group, for optical scene information to be converted to number
Word video image information.
As the preferred embodiment of the present invention, the binocular camera group includes:
First camera module is connected with the video decoding circuit group, for passing through binocular parallax range measurement principle
Obtain the depth map of scene;
Second camera module is connected with the video decoding circuit group, for passing through binocular parallax range measurement principle
Obtain the depth map of scene.
As the preferred embodiment of the present invention, the video decoding circuit group includes:
First video decoding circuit is connected with first camera module, for by first camera
The video data of module acquisition carries out video decoding;
Second video decoding circuit is connected with the second camera module, for by the second camera
The video data of module acquisition carries out video decoding.
As the preferred embodiment of the present invention, the system further includes external interface, with the high-performance treatments
Device is connected, for realizing data transmission.
As the preferred embodiment of the present invention, the external interface includes Ethernet interface and power interface, with
The high-performance processor is connected, for realizing data transmission.
As the preferred embodiment of the present invention, the system further includes FLASH memory, at the high-performance
Reason device is connected, for storing data.
As the preferred embodiment of the present invention, the system further includes DDR memory, with the high-performance treatments
Device is connected, for storing data.
As the preferred embodiment of the present invention, the system further includes CPLD circuit, with the high-performance treatments
Device is connected.
As the preferred embodiment of the present invention, the system further includes RTC circuit, with the high-performance processor
It is connected.
The method that the top-type passenger flow monitor processing based on stereovision technique is realized using above system of the invention,
The method the following steps are included:
(1) Image Acquisition is carried out by binocular camera group, calculates step by the way that preset inside and outside parameter and calibration are three-dimensional
The image that year acquires is corrected;
(2) step is calculated by Stereo matching and generates depth map;
(3) pedestrian detection is carried out by the depth map, and according to the relationship between height and depth map brightness to inspection
The personage of survey is modified;
(4) it determines pedestrian's number, calculates relevant attribute.
As the preferred embodiment of the present invention, the Image Acquisition in the step (1) is adopted including the stage image that tests the speed
Collection and product stage Image Acquisition.
As the preferred embodiment of the present invention, it includes that BM is calculated that the Stereo matching in the step (2), which calculates step,
Step.
As the preferred embodiment of the present invention, the relevant attribute in the step (4) includes average height and position
It sets.
In a specific embodiment of the invention, the invention belongs to real-time passenger flow monitor devices, are related to video data acquiring skill
Art, personnel's detecting and tracking algorithm, real time signal processing Hardware platform design technology and big data technology etc..
Top-type passenger flow monitor device in the present invention is based on stereovision technique and realizes passenger flow statistics and monitoring, effectively advises
The shortcomings that having kept away other passenger flow statistics technologies or passenger flow monitor technology substantially increases the essence that personnel pass in and out automatic identification and statistics
True property.It is compared with traditional technology means, present invention has the advantage that
(1) interference of some non-human factors is effectively excluded
The present invention is according to the circular feature and head and shoulder structure recognition personnel amount of head part and direction, some non-personnel
(such as cart, pet, arm etc.) does not all have the feature on head, therefore above-mentioned interfering object will not be counted into passenger flow statistics number
In.
(2) the highly filtered technology of personnel is realized
Having the elevation information of target customized can be arranged the personnel's height for needing filtration counts later, be provided with
After filtering height, personnel's change lower than the height is no longer counted.Such as children video camera of the height lower than 1.2 meters can not be right
It is counted, so that passenger flow data can be associated with admission ticket income, client's purchasing power etc., has reference significance.
(3) solve the problems, such as that side by side, precision is higher
Infrared emission does not solve the problems, such as that side by side, infrared light curtain is solved the problems, such as side by side by probability, and precision is relative accuracy.And
Passenger flow monitor device in the present invention is installed on the ceiling above entrance, from top to bottom to the disengaging personnel of entrance into
Row imaging, can effectively solve the problems, such as disengaging personnel side by side, precision is higher.
(4) contactless statistics, device installation disguise property is good, does not influence the impression of disengaging personnel
Passenger flow monitor device in the present invention is installed on the ceiling above entrance, good concealment, and disengaging personnel are not
Easily discovery.
Passenger flow monitor device in the present invention is installed on the ceiling above the entrance of closed area, it can realize passenger flow
Statistics, and the number of entry/exit can be distinguished or rest on the number inside some closed area.All entrances in closed area
Passenger flow statistics data are summarized by daemon software, and in conjunction with historical data, real-time passenger flow monitor and early warning may be implemented.
Passenger flow statistics algorithm operates on the high-performance digital signal processor in passenger flow monitor device (TI DSP), carries out
Real-time image analysis.Dsp processor in embedded passenger flow analysis terminal acquires video image in real time, and detects moving target
Characteristic point (feature points), then characteristic point is carried out by the features such as head and shoulder information, distance of center circle poly- on airspace
Matching in class (Cluster) and time domain, to detect and identify pedestrian target.By the real time bidirectional to pedestrian target with
Track may be implemented accurately to count the judgement with direction.These analytic functions are completed inside passenger flow monitor device, do not need to account for
With network bandwidth, network performance is not influenced.
Passenger flow monitor device technique scheme block diagram is shown in Fig. 1, and two, left side camera module forms binocular camera, camera
Data are sent to high-end dsp processor after video decodes, high-end dsp processor simultaneously external DDR FLASH and RTC etc., eventually
End product external interface has Ethernet interface and power interface.
Refering to Figure 1, being related to video data acquiring technology, Ren Yuanjian the invention belongs to real-time passenger flow monitor device
Survey track algorithm, real time signal processing Hardware platform design technology and big data technology etc..It includes camera module, video solution
Code circuit, FLASH memory, DDR memory, CPLD circuit, RTC circuit and high-performance processor, external interface have Ethernet
Interface and power interface.The video decoding circuit is connected with the high-performance processor, the camera module
It is connected with the video decoding circuit, the FLASH memory and DDR memory and the high-performance processor phase
Connection, the CPLD circuit are connected with the high-performance processor, the RTC circuit and the high-performance treatments
Device is connected.
Please refer to Fig. 2, shown in 3, the apparatus structure is made of multiple state display lamps and two cameras.
The high-performance processor is regarded camera by the camera module and the video decoding circuit
Optical scene information in is converted to digital video image information.
The high-performance processor is completed to include but is not limited to depth image generation, target identification, target following and people
The functions such as member's statistical counting, and further the behavior of target person is analyzed on this basis.
The implementation of data-interface include but is not limited to Ethernet, WIFI, 2G 3G 4G etc., data output format includes
But be not limited to location information, trace information, enter and leave directional information, demographic information and original video information etc..
The functional module of people stream counting algorithm in this patent includes five parts: Image Acquisition, image rectification, perspective view
As generating, pedestrian detection, pedestrian tracking is counted.Wherein, pedestrian detection and tracking only calculate in the perspective view of generation.It is detected
Frame is as shown in Figure 4.
1, Image Acquisition: test phase, capture card carries out image data acquiring with binocular camera;The product stage needs
Operation in input module is directly acquired in dsp;
2, by the inside and outside parameter (need to demarcate in advance) of camera, distortion correction and position image rectification: are carried out to the image of acquisition
Shift correction.Correcting algorithm is with reference to calibration stereo algorithm (Bouguet algorithm);The result of correction makes the row pair of left and right two images
Standard, i.e. pole are located at infinite point.
3, Stereo matching: corrected two images are substantially aligned in the horizontal direction, can be given birth to using Stereo Matching Algorithm
At depth map.Three kinds of Stereo Matching Algorithms: BM, SGBM, GC.In view of the requirement of real-time of demographics, using time-consuming most short
BM algorithm.The core concept of BM algorithm is matched in a certain range by search window, as the SAD in search window
When minimum, that is, it is considered match point.Ultimately generate depth map.
4, pedestrian detection: depth map is that a width represents grayscale image of the taken the photograph objects in images from video camera distance, ash
Angle value is bigger, show it is closer from camera, on the contrary, then remoter from camera.In practical application scene, binocular camera is vertically mounted on
Road junction is crossed, the image of shooting is top view, and the perspective view of generation shows as descending from the crown to sole gray value.As a result,
The algorithm for detecting pedestrian is the connected domain detection of top-down layering.In order to determine the number in scene, if different pedestrians
Between be disconnected, then Betti number can consider the number of pedestrian.But under normal circumstances, in especially crowded environment,
It will appear connection below shoulder between men.Therefore when being connected to, we just stop finding pedestrian downwards.And root
According to the relationship between height and depth map brightness, the personage of detection is modified.
We can be judged according to the connected region in disparity map whether pedestrian, under normal circumstances, the head of pedestrian
It is nearest apart from video camera, it is possible that connection below shoulder.Therefore, the number of people is judged simultaneously by carrying out connected domain detection
It is unreliable.Top-down connected domain detection algorithm is used herein, to determine the number of pedestrian, and according to pedestrian's height and depth
Some relationships between the gray scale of figure carry out the amendment of testing result, the former separates when ensure that pedestrian's connection, and the latter is reduced
Pedestrian target is because of division caused by depth matching error and some and inhuman differentiation.The number that pedestrian has been determined it
Afterwards, its attribute, including average height, position etc. are calculated.
Specific algorithm flow is as shown in Figure 5.
The makeover process principle of target detection is according to Fig. 6, and thus figure is it can be concluded that the higher people of height, by
In closer from camera, the ash in depth map is bigger, then the people that height is high, the connected domain area in same layer is greater than short people
Area.Using this principle, the non-human object that the principle is not met in the target that can be will test is filtered.It is counted, is connected to
Two connected domains that the very close and gray value that domain position is leaned on is not much different, are same targets, can merge;For connection
Domain area is not much different, but the different connected domains that gray value but differs greatly rationally are filtered;For connected domain gray value
Larger, position is leaned on close, and but not big enough connected domain merges area.
Using the system and method for the invention for realizing the processing of top-type passenger flow monitor based on stereovision technique, it is used for
Real-time passenger flow statistics and monitoring are effectively eliminated the interference of some non-human factors, are solved simultaneously using 3D rendering identification technology
Row's problem substantially increases the accuracy of passenger flow statistics and monitoring;The present invention realizes the highly filtered technology of personnel, can make by oneself
Justice setting needs personnel's height of filtration counts;The present invention use contactless statistics, device installation disguise property is good, do not influence into
The impression of personnel out.Expection of the present invention can be public in densely populated places such as subway station, airport, market, museum and core scenic spots
Place uses.Can assist manager efficiently, accurately and timely analyze stream of people's situation, reasonable management and distribute public and manpower
Resource promotes public safety, avoids public accident, with intelligent event store function, provides evidence for legal dispute, generates huge
Economic benefit and social benefit.
In this description, the present invention is described with reference to its specific embodiment.But it is clear that can still make
Various modifications and alterations are without departing from the spirit and scope of the invention.Therefore, the description and the appended drawings should be considered as illustrative
And not restrictive.
Claims (13)
1. a kind of system for realizing the processing of top-type passenger flow monitor based on stereovision technique, which is characterized in that the system
Include:
Binocular camera group, for acquiring passenger flow data;
Video decoding circuit group is connected with the binocular camera group, for acquire the binocular camera group
Camera data carry out video decoding;
High-performance processor is connected with the video decoding circuit group, for optical scene information to be converted to digital view
Frequency image information.
2. the system according to claim 1 for realizing the processing of top-type passenger flow monitor based on stereovision technique, feature
It is, the binocular camera group includes:
First camera module is connected with the video decoding circuit group, for being obtained by binocular parallax range measurement principle
The depth map of scene;
Second camera module is connected with the video decoding circuit group, for being obtained by binocular parallax range measurement principle
The depth map of scene.
3. the system according to claim 1 for realizing the processing of top-type passenger flow monitor based on stereovision technique, feature
It is, the video decoding circuit group includes:
First video decoding circuit is connected with first camera module, for by first camera module
The video data of acquisition carries out video decoding;
Second video decoding circuit is connected with the second camera module, for by the second camera module
The video data of acquisition carries out video decoding.
4. the system according to claim 1 for realizing the processing of top-type passenger flow monitor based on stereovision technique, feature
It is, the system further includes external interface, is connected with the high-performance processor, for realizing data transmission.
5. the system according to claim 4 for realizing the processing of top-type passenger flow monitor based on stereovision technique, feature
It is, the external interface includes Ethernet interface and power interface, is connected with the high-performance processor, is used for
Realize data transmission.
6. the system according to claim 1 for realizing the processing of top-type passenger flow monitor based on stereovision technique, feature
It is, the system further includes FLASH memory, is connected with the high-performance processor, for storing data.
7. the system according to claim 1 for realizing the processing of top-type passenger flow monitor based on stereovision technique, feature
It is, the system further includes DDR memory, is connected with the high-performance processor, for storing data.
8. the system according to claim 1 for realizing the processing of top-type passenger flow monitor based on stereovision technique, feature
It is, the system further includes CPLD circuit, is connected with the high-performance processor.
9. the system according to claim 1 for realizing the processing of top-type passenger flow monitor based on stereovision technique, feature
It is, the system further includes RTC circuit, is connected with the high-performance processor.
10. a kind of realize what the top-type passenger flow monitor based on stereovision technique was handled using system described in claim 1
Method, which is characterized in that the method the following steps are included:
(1) Image Acquisition is carried out by binocular camera group, by preset inside and outside parameter and calibration it is three-dimensional calculate step year adopt
The image of collection is corrected;
(2) step is calculated by Stereo matching and generates depth map;
(3) pedestrian detection is carried out by the depth map, and according to the relationship between height and depth map brightness to detection
Personage is modified;
(4) it determines pedestrian's number, calculates relevant attribute.
11. the method for top-type passenger flow monitor processing of the realization according to claim 10 based on stereovision technique,
It is characterized in that, the Image Acquisition in the step (1) includes test the speed stage Image Acquisition and product stage Image Acquisition.
12. the method for top-type passenger flow monitor processing of the realization according to claim 10 based on stereovision technique,
It is characterized in that, it includes that BM calculates step that the Stereo matching in the step (2), which calculates step,.
13. the method for top-type passenger flow monitor processing of the realization according to claim 10 based on stereovision technique,
It is characterized in that, the relevant attribute in the step (4) includes average height and position.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811542849.6A CN109447042A (en) | 2018-12-17 | 2018-12-17 | The system and method for top-type passenger flow monitor processing is realized based on stereovision technique |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811542849.6A CN109447042A (en) | 2018-12-17 | 2018-12-17 | The system and method for top-type passenger flow monitor processing is realized based on stereovision technique |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109447042A true CN109447042A (en) | 2019-03-08 |
Family
ID=65559363
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811542849.6A Pending CN109447042A (en) | 2018-12-17 | 2018-12-17 | The system and method for top-type passenger flow monitor processing is realized based on stereovision technique |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109447042A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111047634A (en) * | 2019-11-13 | 2020-04-21 | 杭州飞步科技有限公司 | Scene depth determination method, device, equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1731456A (en) * | 2005-08-04 | 2006-02-08 | 浙江大学 | Bus passenger traffic statistical method based on stereoscopic vision and system therefor |
CN202584249U (en) * | 2012-03-13 | 2012-12-05 | 毛振刚 | Vehicle-mounted video passenger flow counting system |
CN104504688A (en) * | 2014-12-10 | 2015-04-08 | 上海大学 | Method and system based on binocular stereoscopic vision for passenger flow density estimation |
CN104902258A (en) * | 2015-06-09 | 2015-09-09 | 公安部第三研究所 | Multi-scene pedestrian volume counting method and system based on stereoscopic vision and binocular camera |
CN204929102U (en) * | 2015-09-24 | 2015-12-30 | 深圳市兴发隆智能科技有限公司 | Visitor flow measurement statistical system |
-
2018
- 2018-12-17 CN CN201811542849.6A patent/CN109447042A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1731456A (en) * | 2005-08-04 | 2006-02-08 | 浙江大学 | Bus passenger traffic statistical method based on stereoscopic vision and system therefor |
CN202584249U (en) * | 2012-03-13 | 2012-12-05 | 毛振刚 | Vehicle-mounted video passenger flow counting system |
CN104504688A (en) * | 2014-12-10 | 2015-04-08 | 上海大学 | Method and system based on binocular stereoscopic vision for passenger flow density estimation |
CN104902258A (en) * | 2015-06-09 | 2015-09-09 | 公安部第三研究所 | Multi-scene pedestrian volume counting method and system based on stereoscopic vision and binocular camera |
CN204929102U (en) * | 2015-09-24 | 2015-12-30 | 深圳市兴发隆智能科技有限公司 | Visitor flow measurement statistical system |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111047634A (en) * | 2019-11-13 | 2020-04-21 | 杭州飞步科技有限公司 | Scene depth determination method, device, equipment and storage medium |
CN111047634B (en) * | 2019-11-13 | 2023-08-08 | 杭州飞步科技有限公司 | Scene depth determination method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20190122082A1 (en) | Intelligent content displays | |
CN104303193B (en) | Target classification based on cluster | |
US9489565B2 (en) | Image processing apparatus, image processing method, program, and image processing system | |
CN105631515B (en) | People flow counting system | |
CN104966062B (en) | Video monitoring method and device | |
WO2004042673A2 (en) | Automatic, real time and complete identification of vehicles | |
CN104902258A (en) | Multi-scene pedestrian volume counting method and system based on stereoscopic vision and binocular camera | |
CN102164270A (en) | Intelligent video monitoring method and system capable of exploring abnormal events | |
CN102170563A (en) | Intelligent person capture system and person monitoring management method | |
CN110517501A (en) | A kind of vehicle overload early warning system and method | |
KR20160035121A (en) | Method and Apparatus for Counting Entity by Using Location Information Extracted from Depth Image | |
US12039803B2 (en) | Techniques for automatically identifying secondary objects in a stereo-optical counting system | |
CN103324919A (en) | Video monitoring system based on face recognition and data processing method thereof | |
CN106652291A (en) | Indoor simple monitoring and alarming system and method based on Kinect | |
CN106846297A (en) | Pedestrian's flow quantity detecting system and method based on laser radar | |
CN111950499A (en) | Method for detecting vehicle-mounted personnel statistical information | |
CN115171260A (en) | Intelligent access control system based on face recognition | |
CN202121706U (en) | Intelligent personnel monitoring system | |
Ua-Areemitr et al. | Low-cost road traffic state estimation system using time-spatial image processing | |
US20210133491A1 (en) | System and method for detecting, tracking and counting human objects of interest with an improved height calculation | |
CN109447042A (en) | The system and method for top-type passenger flow monitor processing is realized based on stereovision technique | |
KR20170007070A (en) | Method for visitor access statistics analysis and apparatus for the same | |
CN108280402B (en) | Binocular vision-based passenger flow volume statistical method and system | |
CN209216110U (en) | Support to realize the device of the top-type passenger flow monitor processing based on stereovision technique | |
CN107018359A (en) | A kind of vehicle-carried mobile monitoring system that there is intelligent picture to recognize |
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 |