CN109697392A - Draw the method and device of target object thermodynamic chart - Google Patents

Draw the method and device of target object thermodynamic chart Download PDF

Info

Publication number
CN109697392A
CN109697392A CN201710996058.XA CN201710996058A CN109697392A CN 109697392 A CN109697392 A CN 109697392A CN 201710996058 A CN201710996058 A CN 201710996058A CN 109697392 A CN109697392 A CN 109697392A
Authority
CN
China
Prior art keywords
target
frame
virtual location
thermodynamic chart
location
Prior art date
Application number
CN201710996058.XA
Other languages
Chinese (zh)
Inventor
叶韵
陈宇
武军晖
翁志
Original Assignee
北京京东尚科信息技术有限公司
北京京东世纪贸易有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 北京京东尚科信息技术有限公司, 北京京东世纪贸易有限公司 filed Critical 北京京东尚科信息技术有限公司
Priority to CN201710996058.XA priority Critical patent/CN109697392A/en
Publication of CN109697392A publication Critical patent/CN109697392A/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00624Recognising scenes, i.e. recognition of a whole field of perception; recognising scene-specific objects
    • G06K9/00771Recognising scenes under surveillance, e.g. with Markovian modelling of scene activity
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00335Recognising movements or behaviour, e.g. recognition of gestures, dynamic facial expressions; Lip-reading
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/62Methods or arrangements for recognition using electronic means
    • G06K9/6267Classification techniques
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/18Closed circuit television systems, i.e. systems in which the signal is not broadcast

Abstract

The embodiment of the present invention provides a kind of method and device for drawing target object thermodynamic chart, this method comprises: target object of the detection target video in the start frame of a cycle;Obtain the virtual location of each frame of the target object within the period;The virtual location is mapped as physical location;The thermodynamic chart of the target object is drawn according to the physical location of each frame, can generate thermodynamic chart automatically according to target video, improves the accuracy that thermodynamic chart draws efficiency and thermodynamic chart, and then accurately analyzed target video based on thermodynamic chart.

Description

Draw the method and device of target object thermodynamic chart

Technical field

The present invention relates to field of image recognition and field of computer technology, in particular to a kind of drafting target object The method and device of thermodynamic chart.

Background technique

Currently, many places are both provided with monitor video, video monitoring system is that production is brought convenience, for example, some More under rugged environment, monitor video can replace manpower and carry out monitoring control, and the vision that will not generate people's appearance is tired The physiological defects such as labor guarantee into production safety under normal circumstances.Video monitoring is brought convenience to life, and video monitoring can guarantee to control Peace problem, mounts the system to supermarket, it will be appreciated that indoor situations and customer behavior.

In the implementation of the present invention, inventor has found in the prior art when monitor video analyzes and counts, one Kind is that the target object being directly based upon in video pictures by manpower is counted, this mode lost labor cost and low efficiency Under, it is easy to appear dislocation, and the result counted is also not very accurate.One is the frequencies where statistics light stream, and this mode can To solve the problems, such as large scene, but not consider perspective bring personage aberration problems, causes to analyze inaccuracy to monitor video.

Therefore a kind of method and device for drawing target object thermodynamic chart is needed, accurate thermodynamic chart can be drawn, with base Target video is analyzed in the thermodynamic chart of accurate Drawing.

Above- mentioned information are only used for reinforcing the understanding to background of the invention, therefore it disclosed in the background technology part It may include the information not constituted to the prior art known to persons of ordinary skill in the art.

Summary of the invention

In view of this, the present invention provides a kind of method and device for drawing target object thermodynamic chart, can be based on accurately drawing The thermodynamic chart of system analyzes target video.

Other characteristics and advantages of the invention will be apparent from by the following detailed description, or partially by the present invention Practice and acquistion.

According to the first aspect of the invention, a kind of method for drawing target object thermodynamic chart is provided, wherein the method packet It includes:

Detect target object of the target video in the start frame of a cycle;

Obtain the virtual location of each frame of the target object within the period;

The virtual location is mapped as physical location;

The thermodynamic chart of the target object is drawn according to the physical location of each frame.

According to some embodiments, target object of the detection target video in the start frame of a cycle, comprising:

Based on trained target object detection model mark the target video in the start frame in the period at least One object box, to get the target object in each object box.

According to some embodiments, the virtual location of each frame of the acquisition target object within the period, packet It includes:

Virtual location of the target object in each frame is calculated based on target tracking algorithm.

According to some embodiments, it is described the virtual location is mapped as physical location before, the method also includes:

Region is demarcated in the picture of the target video;

Obtain size of the calibration region in actual scene;

Calculate the picture of the target video and the perspective transformation matrix of actual scene mapping;

It is described that the virtual location is mapped as physical location, comprising:

Virtual location of the target object in each frame is mapped as physical location according to the perspective transformation matrix.

According to some embodiments, described based on target tracking algorithm to calculate the target object virtual in each frame Position, comprising:

Virtual location of the designated key point of the target object in each frame is calculated based on critical point detection algorithm, Using the virtual location of the designated key point as the virtual location of the target object.

According to some embodiments, the thermodynamic chart that the target object is drawn according to the actual path, comprising:

According to the target object in the corresponding physical location of virtual location of each frame, the mesh in the period is calculated Mark object densities and average speed;

According to the target object density and the average speed, the target object is drawn in the actual scene Stop thermodynamic chart.

According to the second aspect of the invention, a kind of device for drawing target object thermodynamic chart is provided, wherein described device packet It includes:

Detection module, for detecting target object of the target video in the start frame of a cycle;

Module is obtained, for obtaining the virtual location of each frame of the target object within the period;

Mapping block, for the virtual location to be mapped as physical location;

Drafting module, for drawing the thermodynamic chart of the target object according to the physical location of each frame.

According to some embodiments, the detection module is configured to described in trained target object detection model label At least one object box of target video in the start frame in the period, to get the target object in each object box.

According to some embodiments, the acquisition module is configured to target tracking algorithm and calculates the target object Virtual location in each frame.

According to some embodiments, described device further include: preprocessing module, for the picture acceptance of the bid in the target video Determine region, obtains size of the calibration region in actual scene, and calculate the picture and reality of the target video The perspective transformation matrix of scene mapping;

The mapping block, for according to the calculated perspective transformation matrix of the preprocessing module by the target Virtual location of the object in each frame is mapped as physical location.

According to some embodiments, the drafting module, comprising:

Computing unit, for according to the target object in each frame virtual location and corresponding physical location, calculate Target object density and average speed in the period out;

Drawing unit, for drawing the target object and existing according to the target object density and the average speed The stop thermodynamic chart of the actual scene.

According to the third aspect of the invention we, a kind of computer readable storage medium is provided, computer program is stored thereon with, Wherein, method and step as described in relation to the first aspect is realized when which is executed by processor.

According to the fourth aspect of the invention, a kind of electronic equipment is provided, wherein include: one or more processors;Storage Device, for storing one or more programs, when one or more of programs are executed by one or more of processors, So that one or more of processors realize method and step as described in relation to the first aspect.

In the embodiment of the present invention, pass through target object of the detection target video in the start frame of a cycle;Obtain institute State the virtual location of each frame of the target object within the period;The virtual location is mapped as physical location;According to every The physical location of one frame draws the thermodynamic chart of the target object, can generate thermodynamic chart automatically according to target video, improve Thermodynamic chart draws the accuracy of efficiency and thermodynamic chart, and then is accurately analyzed based on thermodynamic chart target video.

Detailed description of the invention

Its example embodiment is described in detail by referring to accompanying drawing, above and other target of the invention, feature and advantage will It becomes more fully apparent.

Fig. 1 is a kind of flow chart of method for drawing target object thermodynamic chart shown according to an exemplary embodiment;

Fig. 2 is the schematic diagram that the object box comprising target object is detected in picture shown in the embodiment of the present invention;

Fig. 3 is the schematic diagram after intercepting out by the object box in Fig. 2;

Fig. 4 is the schematic diagram of the virtual location using the intermediate point between people's both feet as people;

Fig. 5 is the schematic diagram that region is demarcated in picture shown in the embodiment of the present invention;

Fig. 6 is a kind of structural representation of device for drawing target object thermodynamic chart shown according to an exemplary embodiment Figure;

Fig. 7 is the structural schematic diagram of a kind of electronic equipment shown according to an exemplary embodiment.

Specific embodiment

Example embodiment is described more fully with reference to the drawings.However, example embodiment can be real in a variety of forms It applies, and is not understood as limited to embodiment set forth herein;On the contrary, thesing embodiments are provided so that the present invention will be comprehensively and complete It is whole, and the design of example embodiment is comprehensively communicated to those skilled in the art.Identical appended drawing reference indicates in figure Same or similar part, thus repetition thereof will be omitted.

In addition, described feature, structure or characteristic can be incorporated in one or more implementations in any suitable manner In example.In the following description, many details are provided to provide and fully understand to the embodiment of the present invention.However, It will be appreciated by persons skilled in the art that technical solution of the present invention can be practiced without one or more in specific detail, Or it can be using other methods, constituent element, device, step etc..In other cases, it is not shown in detail or describes known side Method, device, realization or operation are to avoid fuzzy each aspect of the present invention.

Block diagram shown in the drawings is only functional entity, not necessarily must be corresponding with physically separate entity. I.e., it is possible to realize these functional entitys using software form, or realized in one or more hardware modules or integrated circuit These functional entitys, or these functional entitys are realized in heterogeneous networks and/or processor device and/or microcontroller device.

Flow chart shown in the drawings is merely illustrative, it is not necessary to including all content and operation/step, It is not required to execute by described sequence.For example, some operation/steps can also decompose, and some operation/steps can close And or part merge, therefore the sequence actually executed is possible to change according to the actual situation.

Fig. 1 is a kind of flow chart of method for drawing target object thermodynamic chart shown according to an exemplary embodiment.

As shown in Figure 1, detecting target object of the target video in the start frame of a cycle in S110.

According to example embodiment, target video can be periodically detected, and detects the target video each Target object in the start frame picture in a period.It is small-sized that the target video can include but is not limited to convenience store, small shop etc. The monitor video in place and the monitor video of supermarket, market and large-scale outdoor location.Target object may include but unlimited In people.It should be pointed out that detection cycle can be adaptively adjusted according to target object and application places.For example, in shop When equal places detection customer, since the customer in shop is likely to occur variation, such as new guest at any time, so detection cycle Should be arranged it is shorter, for example, detection in every 2 seconds is primary.

According to example embodiment, it when detecting the target object in start frame, can be detected based on trained target object Model marks at least one object box of the target video in the start frame in the period, to get in each object box Target object.

It should be noted that single-lens detection (Single shot can be passed through in training objective object detection model Detection, SSD) method is trained.

SSD is a kind of convolutional neural networks, and convolutional neural networks difference, which is essentially consisted in, replaces general neural network with convolution In full connection, to obtain response to convolution kernel feature in image in convolution results.Activation primitive is passed through in convolution response The new multichannel response diagram that obtains with pond and to the invariance of micro-displacement, can again pass by the behaviour in convolution activation pond The response diagram for making to obtain higher level feature, after the number of plies enough depths, using response diagram information by full connection transformation or Being directly placed into classifier is exactly sorting algorithm;And it is trained to extract using response diagram information and markup information and is likely to occur mesh The algorithm for marking the region of object is exactly object detection algorithms.

For example, trained target object model can be obtained in the following manner:

1) standard of Pascal VOC receiving will be converted by the region of the obtained multiple pictures of original picture and user's mark Format.

2) it is trained using SSD.

3) after model convergence, obtained model is used to predict in samples pictures.

For example, if actually without target object in samples pictures, but object box can be obtained according to the model, then should Sample retains as the sample of " being easy erroneous detection is target object ".

4) it is added in original picture using misrecognition samples pictures as new one kind, and obtains multiple figures using the picture Piece makes different classes of picture reach quantity balance.

5) detection effect for reaching stable is trained using SSD again.

The only training objective object detection model by taking SSD method as an example, training mesh are pointed out that in the embodiment of the present invention Mark object detection model can also be trained by other methods based on convolutional neural networks.

In S120, the virtual location of each frame of the above-mentioned target object within the period is obtained.

According to example embodiment, the target object can be calculated based on target tracking algorithm virtual in each frame Position.

Target tracking algorithm is used between continuous video frame, according to the position of present frame object to the object of next frame It is tracked, target tracking algorithm is generally divided into two major classes production algorithm and discriminate algorithm.The basic think of of production algorithm Road is modeled to the region currently demarcated, and is then searched for and the most like region of current region in the next frame.Discriminate The thinking of algorithm is trained to current region and periphery inactive area, and classifier is obtained, and then next frame is in target week It encloses and searches for and be made to determine whether to belong to target area.

After start frame detects target object detection, the target object can be calculated based on critical point detection algorithm Virtual location of the designated key point in each frame, using the virtual location of the designated key point as the target object Virtual location.

It should be noted that since the target object is an object for having volume, may therefore specify that the position of a key point Set the virtual location as the target object.For example, when target object is people, using the intermediate point between people's both feet as people's Virtual location.

Artis detection algorithm is used to artis detection algorithm and is used to judge each key of target object in target area The position of point in the zone, such as by taking target object is behaved as an example, key point can be neck, shoulder, knee etc..Critical point detection algorithm It is generally divided into two steps.Step 1: feature extraction is carried out to current region image and is characterized, such as SIFT feature, HOG feature, Harris angle point etc.;Step 2: carrying out pattern-recognition to feature and understanding, so that it is determined that, there is stencil matching in artis position, moves The methods of state planning, machine learning.Also there is neural network method that can combine two steps.It can be any saturating using the algorithm Key point present position is judged in the image including target object optionally taken down.For example, can be with using the algorithm Personage's foot present position is judged in the personage region taken in any perspective, further, by the two of the same person Intermediate point between foot as people present frame virtual location.

In S130, the virtual location is mapped as physical location.

After the virtual location for getting each frame in a cycle, the virtual location of each frame is mapped as actual bit It sets.

It according to example embodiment, can be in the target video before the virtual location is mapped as physical location Picture in demarcate region, obtain size of the calibration region in actual scene, calculate the picture of the target video With the perspective transformation matrix of actual scene mapping.To according to the perspective transformation matrix by the target object in each frame Virtual location is mapped as physical location.

It should be pointed out that the calibration region can be arbitrary size and arbitrary shape.

In S140, the thermodynamic chart of the target object is drawn according to the physical location of each frame.

According to example embodiment, institute is calculated in the corresponding physical location of virtual location of each frame according to target object Target object density and the average speed in the period are stated, and then draws the target object in the stop heat of the actual scene Try hard to.

For example, using actual scene as shop, for target detection object is customer, actual scene can be divided into multiple Unit area, if a unit area of present frame detects target object, customer counts+1.So as to calculate one The sum for the customer that the unit area occurs in all frames in a period, and then the unit area of actual scene is in a week Density per capita in phase:

Sum/frame number/unit area area

It, can be according to actual displacement of the target object between two frame adjacent in the unit area when calculating average speed Divided by the time difference of this two frame, the average speed of this two frame is obtained, and then target object can be calculated in the unit area Per adjacent two frame average speed (in view of target object may move repeatedly in a region, so in the present embodiment The average speed of target object is acquired with the average speed of every adjacent two frame), it, can to being averaging again after the summation of these average speeds To obtain average speed of the target object in the unit area, by all target objects detected in the unit area Average speed summation is average again, the average speed of all target objects in the available unit area within the period:

General speed/customer's counting occur

Further, it is possible to set a maximum speed as Vmax, weeding out actual speed (can greater than the people of the maximum speed To be interpreted as the people only by shop), average speed is subtracted with Vmax, then multiplied by density per capita, obtains stopping heating power Index.

And then visualize obtained result, perspective transform is then carried out by the inverse matrix of perspective transformation matrix, Simple thermodynamic chart is obtained, the thermodynamic chart and original monitored picture are weighted superposition, obtains stopping thermodynamic chart.

In above-described embodiment, comprehensively considers target object density and average speed the two indexs, enrich drafting heating power The index of figure is drawn using the two indexs and stops thermodynamic chart, so that the thermodynamic chart drawn is more accurate.

In the embodiment of the present invention, pass through target object of the detection target video in the start frame of a cycle;Obtain institute State the virtual location of each frame of the target object within the period;The virtual location is mapped as physical location;According to every The physical location of one frame draws the thermodynamic chart of the target object, can generate thermodynamic chart automatically according to target video, improve Thermodynamic chart draws the accuracy of efficiency and thermodynamic chart, and then is accurately analyzed based on thermodynamic chart target video.

For example, in the scenes such as shop, it can be according to heating power map analysis intensity of passenger flow, and then promote to improve, in searching shop Hot spot areas (people is more and speed is slow), puts the commodity for needing to promote on the shelf by the hot spot areas.Population in searching shop Density is low but slow-footed region, maximum probability be laid out it is unreasonable lead to congestion regions, improved based on thermodynamic chart result Improve customer and strolls shop experience.And non-shop scene is equally applicable, in public domain, more particularly, to security protection such as Iron platform, the high fever regional search in the regions such as airport lounge, targetedly reinforces safe strength, prevents the hair for fearing event cruelly It is raw.

The method that target object thermodynamic chart is drawn to one of embodiment of the present invention below with reference to specific application scenarios It is described in detail.Using actual scene as shop in the embodiment, for target detection object is behaved.

Fig. 2 is the schematic diagram that the object box comprising target object is detected in picture shown in the embodiment of the present invention.It needs It is noted that the target object in the present embodiment is customer, it further include other target objects in the picture, Fig. 2 is only with one For object box.After detecting the object box of target object, for the ease of the virtual bit to the target object in the object box Object box can be intercepted out by the calculating set, as shown in figure 3, being the schematic diagram after intercepting out by the object box in Fig. 2. Key point monitoring algorithm is executed to the target object in object box, using the intermediate point between people's both feet as people's in the present embodiment Virtual location, as shown in figure 4, it is the schematic diagram of the virtual location using the intermediate point between people's both feet as people.

Fig. 5 is the schematic diagram that region is demarcated in picture shown in the embodiment of the present invention.The calibration region is rectangle, is obtained It is described calibration region the size in actual scene, such as in the available rectangle any two mutually perpendicular sides reality Border length, and in the calibration region labeling.And then calibration region and the calibration region can be utilized by aberration correction algorithm Actual size, calculate target video picture and actual scene mapping perspective transformation matrix.It is true empty after wherein converting Between in region provide the coordinate at four angles according to (0,0) is followed successively by counterclockwise, (w, 0), (w, h), (0, h) obtains perspective and becomes Change matrix M.

According to physical location corresponding to the intermediate point in the calculated each frame of perspective transformation matrix M between people's bipod, Calculate target object density and the average speed in the period, and then according to the target object density and described flat Equal speed draws the target object in the thermodynamic chart of the actual scene.

It will be clearly understood that the present disclosure describe how being formed and using particular example, but the principle of the present invention is not limited to These exemplary any details.On the contrary, the introduction based on present disclosure, these principles can be applied to many other Embodiment.

Following is apparatus of the present invention embodiment, can be used for executing embodiment of the present invention method.Device is retouched below In stating, part identical with preceding method be will not be described in great detail.

Fig. 6 is a kind of structural representation of device for drawing target object thermodynamic chart shown according to an exemplary embodiment Figure.As shown in fig. 6, described device 600 includes:

Detection module 610, for detecting target object of the target video in the start frame of a cycle;

Module 620 is obtained, for obtaining the virtual location of each frame of the target object within the period;

Mapping block 630, for the virtual location to be mapped as physical location;

Drafting module 640, for drawing the thermodynamic chart of the target object according to the physical location of each frame.

According to some embodiments, the detection module 610 is configured to trained target object detection model label institute At least one object box of target video in the start frame in the period is stated, to get the target pair in each object box As.

According to some embodiments, the acquisition module 620 is configured to target tracking algorithm and calculates the target pair As the virtual location in each frame.

According to some embodiments, described device 600 further include: preprocessing module 650, for the picture in the target video Region is demarcated in face, obtains the size in actual scene in the calibration region, and calculates the picture of the target video The perspective transformation matrix in face and actual scene mapping.

The mapping block 630, for according to the calculated perspective transformation matrix of the preprocessing module 650 by institute It states virtual location of the target object in each frame and is mapped as physical location.

According to some embodiments, the drafting module 640, comprising:

Computing unit 642, for according to the target object in each frame virtual location and corresponding physical location, meter Calculate target object density and the average speed in the period;

Drawing unit 644, for drawing the target object according to the target object density and the average speed In the stop thermodynamic chart of the actual scene.

In the embodiment of the present invention, pass through target object of the detection target video in the start frame of a cycle;Obtain institute State the virtual location of each frame of the target object within the period;The virtual location is mapped as physical location;According to every The physical location of one frame draws the thermodynamic chart of the target object, can generate thermodynamic chart automatically according to target video, improve Thermodynamic chart draws the accuracy of efficiency and thermodynamic chart, and then is accurately analyzed based on thermodynamic chart target video.

As on the other hand, present invention also provides a kind of computer-readable medium, which be can be Included in equipment described in above-described embodiment;It is also possible to individualism, and without in the supplying equipment.Above-mentioned calculating Machine readable medium carries one or more program, when said one or multiple programs are executed by the equipment, makes Obtaining the equipment can execute: target object of the detection target video in the start frame of a cycle;Obtain the target object The virtual location of each frame within the period;The virtual location is mapped as physical location;According to the reality of each frame Draw the thermodynamic chart of the target object in position.

Fig. 7 is the structural schematic diagram of a kind of electronic equipment shown according to an exemplary embodiment.It should be noted that figure Electronic equipment shown in 7 is only an example, should not function to the embodiment of the present application and use scope bring any restrictions.

As shown in fig. 7, computer system 700 includes central processing unit (CPU) 701, it can be read-only according to being stored in Program in memory (ROM) 702 or be loaded into the program in random access storage device (RAM) 703 from storage section 708 and Execute various movements appropriate and processing.In RAM703, also it is stored with system 700 and operates required various programs and data. CPU 701, ROM 702 and RAM 703 are connected with each other by bus 704.Input/output (I/O) interface 705 is also connected to always Line 704.

I/O interface 705 is connected to lower component: the importation 706 including keyboard, mouse etc.;It is penetrated including such as cathode The output par, c 707 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.;Storage section 708 including hard disk etc.; And the communications portion 709 of the network interface card including LAN card, modem etc..Communications portion 709 via such as because The network of spy's net executes communication process.Driver 710 is also connected to I/O interface 705 as needed.Detachable media 711, such as Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 710, in order to read from thereon Computer program be mounted into storage section 708 as needed.

Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium On computer program, which includes the program code for method shown in execution flow chart.In such reality It applies in example, which can be downloaded and installed from network by communications portion 709, and/or from detachable media 711 are mounted.When the computer program is executed by central processing unit (CPU) 701, executes and limited in the terminal of the application Above-mentioned function.

It should be noted that computer-readable medium shown in the application can be computer-readable signal media or meter Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device, Or above-mentioned any appropriate combination.In this application, computer readable storage medium can be it is any include or storage journey The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this In application, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal, Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium Program code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned Any appropriate combination.

Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the application, method and computer journey The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants It is noted that the combination of each box in block diagram or flow chart and the box in block diagram or flow chart, can use and execute rule The dedicated hardware based systems of fixed functions or operations is realized, or can use the group of specialized hardware and computer instruction It closes to realize.

Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard The mode of part is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor packet It includes detection module, obtain module, mapping block and drafting module.Wherein, the title of these modules not structure under certain conditions The restriction of the pairs of module itself.

Specifically illustrate and describe above exemplary embodiment of the present invention.It should be appreciated that the present invention is not limited to this In the detailed construction, set-up mode or the implementation method that describe;On the contrary, it is intended to cover the essence included in appended claims Various modifications and equivalence setting in mind and range.

Claims (13)

1. a kind of method for drawing target object thermodynamic chart, which is characterized in that the described method includes:
Detect target object of the target video in the start frame of a cycle;
Obtain the virtual location of each frame of the target object within the period;
The virtual location is mapped as physical location;
The thermodynamic chart of the target object is drawn according to the physical location of each frame.
2. the method as described in claim 1, which is characterized in that the detection target video is in the start frame of a cycle Target object, comprising:
At least one of the target video in the start frame in the period is marked based on trained target object detection model Object box, to get the target object in each object box.
3. the method as described in claim 1, which is characterized in that the acquisition target object is each within the period The virtual location of frame, comprising:
Virtual location of the target object in each frame is calculated based on target tracking algorithm.
4. the method as described in claim 1, which is characterized in that it is described the virtual location is mapped as physical location before, The method also includes:
Region is demarcated in the picture of the target video;
Obtain size of the calibration region in actual scene;
Calculate the picture of the target video and the perspective transformation matrix of actual scene mapping;
It is described that the virtual location is mapped as physical location, comprising:
Virtual location of the target object in each frame is mapped as physical location according to the perspective transformation matrix.
5. method as claimed in claim 3, which is characterized in that described to calculate the target object based on target tracking algorithm Virtual location in each frame, comprising:
Virtual location of the designated key point of the target object in each frame is calculated based on critical point detection algorithm, by institute State virtual location of the virtual location of designated key point as the target object.
6. the method as described in claim 1, which is characterized in that described to draw the target object according to the actual path Thermodynamic chart, comprising:
According to the target object in the corresponding physical location of virtual location of each frame, the target pair in the period is calculated As density and average speed;
According to the target object density and the average speed, the target object is drawn in the stop of the actual scene Thermodynamic chart.
7. a kind of device for drawing target object thermodynamic chart, which is characterized in that described device includes:
Detection module, for detecting target object of the target video in the start frame of a cycle;
Module is obtained, for obtaining the virtual location of each frame of the target object within the period;
Mapping block, for the virtual location to be mapped as physical location;
Drafting module, for drawing the thermodynamic chart of the target object according to the physical location of each frame.
8. device as claimed in claim 7, which is characterized in that the detection module is configured to trained target object Detection model marks at least one object box of the target video in the start frame in the period, to get each object Target object in frame.
9. device as claimed in claim 7, which is characterized in that the acquisition module is configured to target tracking algorithm meter Calculate virtual location of the target object in each frame.
10. device as claimed in claim 7, which is characterized in that described device further include: preprocessing module, for described Region is demarcated in the picture of target video, obtains size of the calibration region in actual scene, and calculate the mesh Mark the picture of video and the perspective transformation matrix of actual scene mapping;
The mapping block, for according to the calculated perspective transformation matrix of the preprocessing module by the target object Virtual location in each frame is mapped as physical location.
11. device as claimed in claim 7, which is characterized in that the drafting module, comprising:
Computing unit, for, in the corresponding physical location of virtual location of each frame, being calculated described according to the target object Target object density and average speed in period;
Drawing unit, for drawing the target object described according to the target object density and the average speed The stop thermodynamic chart of actual scene.
12. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor Method and step described in any one of claims 1-6 is realized when execution.
13. a kind of electronic equipment characterized by comprising one or more processors;
Storage device, for storing one or more programs, when one or more of programs are by one or more of processing When device executes, so that one or more of processors realize such as method and step of any of claims 1-6.
CN201710996058.XA 2017-10-23 2017-10-23 Draw the method and device of target object thermodynamic chart CN109697392A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710996058.XA CN109697392A (en) 2017-10-23 2017-10-23 Draw the method and device of target object thermodynamic chart

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710996058.XA CN109697392A (en) 2017-10-23 2017-10-23 Draw the method and device of target object thermodynamic chart

Publications (1)

Publication Number Publication Date
CN109697392A true CN109697392A (en) 2019-04-30

Family

ID=66226174

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710996058.XA CN109697392A (en) 2017-10-23 2017-10-23 Draw the method and device of target object thermodynamic chart

Country Status (1)

Country Link
CN (1) CN109697392A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020233221A1 (en) * 2019-05-22 2020-11-26 北京京东尚科信息技术有限公司 Heatmap analysis method, device, and system

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102385690A (en) * 2010-09-01 2012-03-21 汉王科技股份有限公司 Target tracking method and system based on video image
CN102629385A (en) * 2012-02-28 2012-08-08 中山大学 Object matching and tracking system based on multiple camera information fusion and method thereof
CN102830793A (en) * 2011-06-16 2012-12-19 北京三星通信技术研究有限公司 Sight tracking method and sight tracking device
CN103607554A (en) * 2013-10-21 2014-02-26 无锡易视腾科技有限公司 Fully-automatic face seamless synthesis-based video synthesis method
CN104915970A (en) * 2015-06-12 2015-09-16 南京邮电大学 Multi-target tracking method based on track association
CN105447458A (en) * 2015-11-17 2016-03-30 深圳市商汤科技有限公司 Large scale crowd video analysis system and method thereof
CN105959382A (en) * 2016-06-06 2016-09-21 上海论宇信息科技有限公司 Intelligent movement monitoring device and monitoring method
CN106023252A (en) * 2016-05-16 2016-10-12 浙江理工大学 Multi-camera human body tracking method based on OAB algorithm
CN106339690A (en) * 2016-08-30 2017-01-18 上海交通大学 Video object flow detecting method and system based on noise elimination and auxiliary determination line
CN106504270A (en) * 2016-11-08 2017-03-15 浙江大华技术股份有限公司 The methods of exhibiting and device of target object in a kind of video
CN107103279A (en) * 2017-03-09 2017-08-29 广东顺德中山大学卡内基梅隆大学国际联合研究院 A kind of passenger flow counting method under vertical angle of view based on deep learning

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102385690A (en) * 2010-09-01 2012-03-21 汉王科技股份有限公司 Target tracking method and system based on video image
CN102830793A (en) * 2011-06-16 2012-12-19 北京三星通信技术研究有限公司 Sight tracking method and sight tracking device
CN102629385A (en) * 2012-02-28 2012-08-08 中山大学 Object matching and tracking system based on multiple camera information fusion and method thereof
CN103607554A (en) * 2013-10-21 2014-02-26 无锡易视腾科技有限公司 Fully-automatic face seamless synthesis-based video synthesis method
CN104915970A (en) * 2015-06-12 2015-09-16 南京邮电大学 Multi-target tracking method based on track association
CN105447458A (en) * 2015-11-17 2016-03-30 深圳市商汤科技有限公司 Large scale crowd video analysis system and method thereof
CN106023252A (en) * 2016-05-16 2016-10-12 浙江理工大学 Multi-camera human body tracking method based on OAB algorithm
CN105959382A (en) * 2016-06-06 2016-09-21 上海论宇信息科技有限公司 Intelligent movement monitoring device and monitoring method
CN106339690A (en) * 2016-08-30 2017-01-18 上海交通大学 Video object flow detecting method and system based on noise elimination and auxiliary determination line
CN106504270A (en) * 2016-11-08 2017-03-15 浙江大华技术股份有限公司 The methods of exhibiting and device of target object in a kind of video
CN107103279A (en) * 2017-03-09 2017-08-29 广东顺德中山大学卡内基梅隆大学国际联合研究院 A kind of passenger flow counting method under vertical angle of view based on deep learning

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
苟超: ""室内场景的人体跟踪与行为分析"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020233221A1 (en) * 2019-05-22 2020-11-26 北京京东尚科信息技术有限公司 Heatmap analysis method, device, and system

Similar Documents

Publication Publication Date Title
Muhammad et al. Efficient deep CNN-based fire detection and localization in video surveillance applications
US9953217B2 (en) System and method for pose-aware feature learning
Zhang et al. Deep convolutional neural networks for forest fire detection
KR101768521B1 (en) Method and system providing informational data of object included in image
Seer et al. Kinects and human kinetics: A new approach for studying pedestrian behavior
Bertini et al. Multi-scale and real-time non-parametric approach for anomaly detection and localization
US9098740B2 (en) Apparatus, method, and medium detecting object pose
CN105353634B (en) Utilize the home appliance and method of gesture identification control operation
Arietta et al. City forensics: Using visual elements to predict non-visual city attributes
US8582811B2 (en) Unsupervised parameter settings for object tracking algorithms
Li et al. Anomaly detection and localization in crowded scenes
CN105122270B (en) The method and system of people is counted using depth transducer
Idrees et al. Detecting humans in dense crowds using locally-consistent scale prior and global occlusion reasoning
US20170372164A1 (en) Object detection and classification
US9460518B2 (en) Visual clothing retrieval
US9928423B2 (en) Efficient retrieval of anomalous events with priority learning
CN105373785B (en) Gesture identification detection method and device based on deep neural network
JP5647627B2 (en) Unusual pattern discovery
CN104885111B (en) Control method for displaying sales information on information terminal
JP2013206462A (en) Method for measuring parking lot occupancy state from digital camera image
US8983133B2 (en) Multi-view object detection using appearance model transfer from similar scenes
KR20200037833A (en) Item detection system and item detection method based on image monitoring
US7995841B2 (en) Hybrid graph model for unsupervised object segmentation
Greiffenhagen et al. Design, analysis, and engineering of video monitoring systems: an approach and a case study
CN104572804B (en) A kind of method and its system of video object retrieval

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