CN109448026A - Passenger flow statistical method and system based on head and shoulder detection - Google Patents
Passenger flow statistical method and system based on head and shoulder detection Download PDFInfo
- Publication number
- CN109448026A CN109448026A CN201811367834.0A CN201811367834A CN109448026A CN 109448026 A CN109448026 A CN 109448026A CN 201811367834 A CN201811367834 A CN 201811367834A CN 109448026 A CN109448026 A CN 109448026A
- Authority
- CN
- China
- Prior art keywords
- head
- shoulder
- shop
- tracking
- personnel
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30242—Counting objects in image
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of passenger flow statistical method based on head and shoulder detection, comprising: S1: the video image in one specified region of acquisition extracts the head and shoulder information of each personnel in video image;S2: successively analyzing each head and shoulder tracking ID, records the corresponding all head and shoulder coordinates of non-salesman user's head and shoulder tracking ID until non-salesman user leaves the specified region, to obtain the movement track of non-salesman user;S3: whether the movement track that the personnel of the non-salesman in specified region have been left in judgement constitutes the behavior of disengaging shop, if so, passenger flow counting adds 1, if not, passenger flow counting adds 0;S4: stopping tracking and the head and shoulder tracking ID is deleted from monitoring list.The present invention judges the disengaging shop behavior of personnel using head and shoulder detection technique, excludes salesman using face recognition technology, realizes the accurate passenger flow statistics in shop.
Description
Technical field
The present invention relates to passenger flow statistics fields, in particular to a kind of passenger flow statistical method based on head and shoulder detection and are
System.
Background technique
Head and shoulder detection technique refers to based on image processing techniques, for the video frame images of input, can detecte video
The tracking ID of head and shoulder, the coordinate position of head and shoulder in head and shoulder and output video frame in frame.Head and shoulder detection mentioned in the present invention
SDK realizes video frame as input, exports as the tracking ID of head and shoulders all in video frame, the coordinate position of head and shoulder.Video
The corresponding head and shoulder of a people in frame.
Human face detection tech refers to based on image processing techniques, for the video frame images of input, can detecte face
And export the position of face.Face datection SDK mentioned in the present invention realizes video frame as input, exports as video
The coordinate of all faces in frame.
Recognition of face, which refers to, carries out characteristics extraction to facial image, and compares with other face characteristic values, takes highest
Alignment score, if highest alignment score be more than setting threshold value, then it is assumed that two faces are the same persons.Institute in the present invention
The characteristics extraction SDK mentioned realizes facial image as input, exports the characteristic value for facial image.
Passenger flow statistics refer to using passenger flow statistics equipment, the volume of the flow of passengers of each period in the shop Lai Tongji.Realize passenger flow
Statistical method has very much, including manually counting, infrared induction, RFID signal acquisition, the passenger flow statistics based on image processing techniques.
Face recognition technology in image processing techniques can exclude salesman in passenger flow statistics, so that the data of passenger flow statistics are more
Precisely.
Summary of the invention
It is an object of that present invention to provide a kind of passenger flow statistical methods and system based on head and shoulder detection, detect skill using head and shoulder
Art judges the disengaging shop behavior of personnel, excludes salesman using face recognition technology, realizes the accurate passenger flow statistics in shop.
To reach above-mentioned purpose, a kind of passenger flow statistical method based on head and shoulder detection, which comprises
S1: the video image in one specified region of acquisition extracts the head and shoulder information of each personnel in video image, head and shoulder information
ID, head and shoulder coordinate are tracked including at least head and shoulder, head and shoulder tracking ID is stored to a monitoring list;
S2: successively analyzing each head and shoulder tracking ID, records the corresponding all head and shoulder coordinates of non-salesman user's head and shoulder tracking ID
Until non-salesman user leaves the specified region, to obtain the movement track of non-salesman user;
S3: whether the movement track that the personnel of the non-salesman in specified region have been left in judgement constitutes the behavior of disengaging shop, if
It is that passenger flow counting adds 1, if not, passenger flow counting adds 0;
S4: stopping tracking and the head and shoulder tracking ID is deleted from monitoring list;
S5: shop interior zone is divided by several sub-regions according to item property, analysis is described to have left specified area
The movement track of the personnel of the non-salesman in domain counts the personnel in the residence time of each sub-regions, according to the rule of a setting
For the personnel, corresponding consumption habit label is set.
Further, in step 2, successively analyze whether the corresponding personnel of each head and shoulder tracking ID are salesman, if so,
The identity attribute of head and shoulder tracking ID is labeled as salesman until the personnel corresponding to it leave the specified area in monitoring list
Domain if not, the identity attribute of head and shoulder tracking ID is labeled as non-salesman in monitoring list, while recording head and shoulder tracking
ID all head and shoulder coordinates are until the personnel corresponding to it leave the specified region, to obtain the movement track of the personnel.
Further, whether the corresponding personnel of each head and shoulder tracking ID of the analysis are that salesman's method includes:
It obtains head and shoulder and tracks ID, it is compared with the head and shoulder tracking ID in monitoring list to obtain its identity attribute, with
And
If head and shoulder tracking ID is not present in monitoring list or is present in monitoring list but do not do identity attribute mark
Note obtains the head and shoulder tracking face characteristic value corresponding to ID, the face characteristic value that will acquire and salesman's face characteristic value number
It compares according to library, to judge whether it is salesman, if so, head and shoulder tracking ID is labeled as salesman in monitoring list, it is no
Then, head and shoulder tracking ID is labeled as non-salesman in monitoring list.
Further, the method also includes:
By the decoding video stream of the video image in the specified region at several video frames;
Head and shoulder detection and analysis, Face datection analysis, face characteristic value extraction and analysis are executed to each video frame, to obtain view
The head and shoulder of each personnel tracks ID, head and shoulder coordinate, face coordinate, face characteristic value information in frequency frame;
The head and shoulder tracking ID that will acquire is stored to monitoring list, is establishd or updated head and shoulder tracking ID head and shoulder corresponding with its and is sat
The mapping relations of mark, face coordinate, face characteristic value information.
Further, the head and shoulder tracking ID that will acquire is stored to monitoring list, while establising or updating head and shoulder tracking
ID and corresponding head and shoulder coordinate, face coordinate, face characteristic value information mapping relations refer to,
The head and shoulder tracking ID that will acquire compares with head and shoulder tracking ID existing in monitoring list, and 1) if the head and shoulder is tracked
ID is not present in monitoring list, is stored to monitoring list, establishes corresponding head and shoulder coordinate database, face coordinate
Database, face characteristic value information database store head and shoulder coordinate, face coordinate, face characteristic value information respectively to head and shoulder
Coordinate database, face coordinate database, in face characteristic value information database, 2) if head and shoulder tracking ID is present in monitoring
In list, head and shoulder coordinate, face coordinate, face characteristic value information are stored respectively to head and shoulder corresponding to head and shoulder tracking ID
Coordinate database, face coordinate database, in face characteristic value information database.
Further, whether the movement track of the personnel of the non-salesman for judging to have left specified region constitutes disengaging shop
The method of behavior includes:
Action trail is taken turns doing into shop logic judgment and shop logic judgment out, will be met simultaneously into shop logic judgment standard
The action trail of shop logic judgment standard is demarcated as constituting the behavior of disengaging shop out.
Further, the method into shop logic judgment includes:
By specified region division at shop region, shop doorway region, shop interior zone outdoors;
Action trail is analyzed, the action trail for meeting the following conditions simultaneously is judged to meeting into shop logic judgment standard:
1) starting point of action trail is in shop outdoors regional scope, 2) terminal of action trail is within the scope of the interior zone of shop,
3) more at least it is present in action trail in the regional scope of shop doorway, otherwise, it is determined that action trail is unsatisfactory for patrolling into shop
Collect judgement.
Further, the method for the shop logic judgment out includes:
By specified region division at shop region, shop doorway region, shop interior zone outdoors;
Action trail is analyzed, the action trail for meeting the following conditions simultaneously is judged to meeting shop logic judgment standard out:
1) starting point of action trail is within the scope of the interior zone of shop, 2) terminal of action trail is in shop outdoors regional scope,
3) more at least it is present in action trail in the regional scope of shop doorway, is patrolled otherwise, it is determined that action trail is unsatisfactory for out shop
Collect judgement.
Further, the method also includes:
Shop interior zone is divided into several sub-regions according to item property, specified region has each been left in analysis
The movement track of the personnel of non-salesman, the personnel that each non-salesman for having left specified region is calculated are funny in each sub-regions
The time stayed;
The total time t that the personnel of all non-salesmans in each subregion within the scope of a setting time stay is counted, in conjunction with same
Within the scope of one setting time in each subregion commodity sale number h, each subregion commodity are calculated according to following formula
Fast-selling degree S within the scope of this setting time:
S=ω1×t+ω2×h
Wherein, ω1、ω2The weight factor of respectively total time t and sale number h, ω1+ω2=1.
On the basis of preceding method, the present invention further mentions a kind of passenger flow statistical system based on head and shoulder detection, the system
System includes:
Video acquisition device, to acquire the video image in a specified region, the specified region includes shop area outdoors
Domain, shop doorway region, shop interior zone;
Video parsing module, to extract the head and shoulder information of each personnel in video image, head and shoulder information includes at least head
Shoulder tracks ID, head and shoulder coordinate, and head and shoulder tracking ID is stored to a monitoring list;
Whether personnel identity determination module is salesman to analyze the corresponding personnel of each head and shoulder tracking ID;
Action trail integrates module, the head and shoulder coordinate of each head and shoulder tracking ID is integrated into action trail;
Disengaging shop behavior determination module, to determine whether the action trail of personnel constitutes the behavior of disengaging shop;
Counting module, to count passenger flow counting;
Label setup module analyzes institute shop interior zone is divided into several sub-regions according to item property
The movement track for having left the personnel of non-salesman in specified region is stated, the personnel is counted in the residence time of each sub-regions, presses
Corresponding consumption habit label is arranged as the personnel in the rule set according to one.
Further, the video parsing module includes:
Video decoding sub-module, to by the decoding video stream of the video image in the specified region at several video frames
Module;
Head and shoulder tests and analyzes submodule, extracted corresponding to head and shoulder tracking ID, each head and shoulder tracking ID from video frame
Head and shoulder coordinate module;
Face datection analyzes submodule, to extract face coordinate corresponding to each head and shoulder tracking ID from video frame
Module;
Face characteristic value extraction and analysis submodule, to extract face corresponding to each head and shoulder tracking ID from video frame
The module of characteristic value.
By the above technical solution of the present invention, compared with existing, significant beneficial effect is,
1) movement track that personnel are obtained using head and shoulder detection technique, by judging whether action trail meets disengaging shop row
For precisely to count passenger flow number.
2) user identity can quickly be identified using face recognition technology.
It should be appreciated that as long as aforementioned concepts and all combinations additionally conceived described in greater detail below are at this
It can be viewed as a part of the subject matter of the disclosure in the case that the design of sample is not conflicting.In addition, required guarantor
All combinations of the theme of shield are considered as a part of the subject matter of the disclosure.
Can be more fully appreciated from the following description in conjunction with attached drawing present invention teach that the foregoing and other aspects, reality
Apply example and feature.The features and/or benefits of other additional aspects such as illustrative embodiments of the invention will be below
Description in it is obvious, or learnt in practice by the specific embodiment instructed according to the present invention.
Detailed description of the invention
Attached drawing is not intended to drawn to scale.In the accompanying drawings, identical or nearly identical group each of is shown in each figure
It can be indicated by the same numeral at part.For clarity, in each figure, not each component part is labeled.
Now, example will be passed through and the embodiments of various aspects of the invention is described in reference to the drawings, in which:
Fig. 1 is the passenger flow statistical method flow chart of the invention based on head and shoulder detection.
Fig. 2 is the method flow diagram of one of example of the invention.
Fig. 3 is satisfaction of the invention into the schematic diagram of the movement track one of shop logic judgment.
Fig. 4 is the schematic diagram into the movement track two of shop logic judgment that is unsatisfactory for of the invention.
Fig. 5 is the schematic diagram into the movement track three of shop logic judgment that is unsatisfactory for of the invention.
Specific embodiment
In order to better understand the technical content of the present invention, special to lift specific embodiment and institute's accompanying drawings is cooperated to be described as follows.
Various aspects with reference to the accompanying drawings to describe the present invention in the disclosure, shown in the drawings of the embodiment of many explanations.
It is not intended to cover all aspects of the invention for embodiment of the disclosure.It should be appreciated that a variety of designs and reality presented hereinbefore
Those of apply example, and describe in more detail below design and embodiment can in many ways in any one come it is real
It applies, this is because conception and embodiment disclosed in this invention are not limited to any embodiment.In addition, disclosed by the invention one
A little aspects can be used alone, or otherwise any appropriately combined use with disclosed by the invention.
In conjunction with Fig. 1, the purpose of the present invention is to propose to a kind of passenger flow statistical method based on head and shoulder detection, the method includes
Following steps:
S1: the video image in one specified region of acquisition extracts the head and shoulder information of each personnel in video image, head and shoulder information
ID, head and shoulder coordinate are tracked including at least head and shoulder, head and shoulder tracking ID is stored to a monitoring list.
S2: successively analyzing each head and shoulder tracking ID, records the corresponding all head and shoulder coordinates of non-salesman user's head and shoulder tracking ID
Until non-salesman user leaves the specified region, to obtain the movement track of non-salesman user.
S3: whether the movement track that the personnel of the non-salesman in specified region have been left in judgement constitutes the behavior of disengaging shop, if
It is that passenger flow counting adds 1, if not, passenger flow counting adds 0.
S4: stopping tracking and the head and shoulder tracking ID is deleted from monitoring list;
S5: shop interior zone is divided by several sub-regions according to item property, analysis is described to have left specified area
The movement track of the personnel of the non-salesman in domain counts the personnel in the residence time of each sub-regions, according to the rule of a setting
For the personnel, corresponding consumption habit label is set.
In conjunction with Fig. 2, for example, installing video camera on shop doorway, captured in real-time one specifies the video image in region, preferably
, it is the disengaging shop in step S3 that this specified region, which includes shop region, shop doorway region, shop interior zone outdoors,
Behavior judgement provides basis.
In step S1, first by video camera shooting decoding video stream be video frame, such as by the decoding video stream of 1s at
25 frame video frames, the video frame frame number that the decoding video stream in the unit time generates is more, and the movement track recorded is more smart
Really, to disengaging shop user's recognition speed faster, the response time is shorter.
On this basis, successively it is used as head and shoulder detection SDK, Face datection SDK, characteristic value to mention the video frame that decoding obtains
The input of SDK is taken, output information is obtained.Head and shoulder detect SDK output video frame in each head and shoulder tracking ID (i.e. TrackID),
Head and shoulder coordinate.The coordinate of the tracking ID, the corresponding face of tracking ID of each head and shoulder in Face datection SDK output video frame.Feature
Value extracts the characteristic value that SDK exports the tracking ID of each head and shoulder, the corresponding face of tracking ID.The same head of three SDK output
The TrackID of shoulder is consistent.
The head and shoulder tracking ID that will acquire is stored to monitoring list, is establishd or updated head and shoulder tracking ID head and shoulder corresponding with its and is sat
The mapping relations of mark, face coordinate, face characteristic value information.
In some instances, the method for establising or updating such mapping relations is as follows:
The head and shoulder tracking ID that will acquire compares with head and shoulder tracking ID existing in monitoring list, and 1) if the head and shoulder is tracked
ID is not present in monitoring list, is stored to monitoring list, establishes corresponding head and shoulder coordinate database, face coordinate
Database, face characteristic value information database store head and shoulder coordinate, face coordinate, face characteristic value information respectively to head and shoulder
Coordinate database, face coordinate database, in face characteristic value information database, 2) if head and shoulder tracking ID is present in monitoring
In list, head and shoulder coordinate, face coordinate, face characteristic value information are stored respectively to head and shoulder corresponding to head and shoulder tracking ID
Coordinate database, face coordinate database, in face characteristic value information database.
The present invention proposes one of them example specifically executed about step S2, specific as follows:
Successively analyze whether the corresponding personnel of each head and shoulder tracking ID are salesman, if so, by the head in monitoring list
The identity attribute of shoulder tracking ID is labeled as salesman until the personnel corresponding to it leave the specified region, if not, monitoring
The identity attribute of head and shoulder tracking ID is labeled as non-salesman in list, while recording all head and shoulder coordinates of head and shoulder tracking ID
Until the personnel corresponding to it leave the specified region, to obtain the movement track of the personnel.
Further, whether the corresponding personnel of each head and shoulder tracking ID of the analysis are that salesman's method includes:
It obtains head and shoulder and tracks ID, it is compared with the head and shoulder tracking ID in monitoring list to obtain its identity attribute, with
And it if head and shoulder tracking ID is not present in monitoring list or is present in monitoring list but do not do identity attribute label, obtains
The head and shoulder is tracked into face characteristic value corresponding to ID, the face characteristic value that will acquire and salesman's face characteristic value database do ratio
It is right, to judge whether it is salesman, if so, in monitoring list otherwise head and shoulder tracking ID is being monitored labeled as salesman
Head and shoulder tracking ID is labeled as non-salesman in list.
Specifically, judging whether the one of TrackID obtained has determined that identity, i.e. personnel corresponding to it are salesmans
Or non-salesman.If having judged, the identity of counterpart personnel for salesman, terminates for this TrackID in the analysis of current video frame,
Continue to analyze next TrackID in current video frame.If the identity of counterpart personnel unconfirmed is salesman, continue to judge this
Whether TrackID has corresponding face, if there is face, obtains the characteristic value of face, does ratio with the face database characteristic value of salesman
It is right, judge whether it is salesman.
Meanwhile all TrackID in each video frame are analyzed, in current TrackID monitoring list
TrackID is compared, if TrackID present in monitoring list does not occur in the video frame, illustrates that this TrackID leaves,
S3 is entered step, judges whether the track of this TrackID meets disengaging shop condition.If satisfied, judge again this TrackID whether be
Salesman, if not salesman, then the appearance of this TrackID to the disengaging shop behavior left as a customer, passenger flow counting add 1.
This TrackID is deleted from monitoring list, process terminates.
As businessman, other than passenger flow quantity statistics, it is also desirable to obtain the consumption habit into customer family, on the one hand, sentence
On the other hand disconnected shop commodity fast sale degree is provided for client and is targetedly serviced.
On this basis, the passenger flow statistical method based on head and shoulder detection mentioned by the present invention further includes following step:
Shop interior zone is divided into several sub-regions according to item property, analysis is described to have left specified region
The movement track of the personnel of non-salesman counts the personnel in the residence time of each sub-regions, and the rule set according to one is is somebody's turn to do
Corresponding consumption habit label is arranged in personnel.
By taking a supermarket as an example, the volume of the flow of passengers is big, items for merchandising type is more, advertising campaign is frequent, is based on preceding method, can incite somebody to action
Supermarket's interior zone is divided into daily necessities region, fresh region, snacks region, drinks region, clothes region according to item property
Etc., according to preceding method, when judgement, any one client leaves the supermarket, analyzes its movement track inside supermarket, counts
Calculation obtains the client in the time that daily necessities region, fresh region, snacks region, drinks region, clothes region are stayed and is respectively
20 minutes, 10 minutes, 5 minutes, 3 minutes and 0 minute.
It is as follows that we set label setting rule: arranging each sub-regions sequence according to duration is descending, before selection
The item property classification of two sub-regions is as this consumption habit label of the client.
For aforementioned client, after its departure, consumption habit label is set as daily necessities, fresh.
This method is of great significance for maintenance member client, and trade company can be according to the consumption habit mark of the member client
Label are increased customer satisfaction degree to targetedly servicing.
It should be appreciated that label setting rule be not limited to it is aforementioned this is a kind of, trade company can self-setting according to demand.
Further, the method also includes:
Shop interior zone is divided into several sub-regions according to item property, specified region has each been left in analysis
The movement track of the personnel of non-salesman, the personnel that each non-salesman for having left specified region is calculated are funny in each sub-regions
The time stayed.
The total time t that the personnel of all non-salesmans in each subregion within the scope of a setting time stay is counted, in conjunction with same
Within the scope of one setting time in each subregion commodity sale number h, each subregion commodity are calculated according to following formula
Fast-selling degree S within the scope of this setting time:
S=ω1×t+ω2×h
Wherein, ω1、ω2The weight factor of respectively total time t and sale number h, ω1+ω2=1.
Preceding method can help businessman accurately to know the fast-selling degree of merchandising, provide data for the commercial analysis of businessman
It supports.
The present invention provides one of judgement left the personnel of the non-salesman in specified region movement track whether structure
At the method for disengaging shop behavior, comprising:
Action trail is taken turns doing into shop logic judgment and shop logic judgment out, will be met simultaneously into shop logic judgment standard
The action trail of shop logic judgment standard is demarcated as constituting the behavior of disengaging shop out.
Wherein, the method into shop logic judgment includes:
By specified region division at shop region, shop doorway region, shop interior zone outdoors.
Action trail is analyzed, the action trail for meeting the following conditions simultaneously is judged to meeting into shop logic judgment standard:
1) starting point of action trail is in shop outdoors regional scope, 2) terminal of action trail is within the scope of the interior zone of shop,
3) more at least it is present in action trail in the regional scope of shop doorway, otherwise, it is determined that action trail is unsatisfactory for patrolling into shop
Collect judgement.
Fig. 3 is the schematic diagram for meeting the movement track into shop logic judgment, and Fig. 4, Fig. 5 are to be unsatisfactory for into shop logic judgment
Movement track schematic diagram.
In Fig. 3, Fig. 4, Fig. 5, the region A of straight line a or more indicates shop region outdoors, and the region B of rectangle b covering indicates shop
Spread doorway region, rectangle b region C below indicates shop interior zone, the pattern of the region D of the two sides rectangle b in conventional shop
In laying, it is not belonging to current shop range.
1) movement track one
The starting point 1-1 of the movement track one of client 1 is located at region A, and terminal 1-3 is located at region C, movement track one its
In an intermediate point 1-2 be located at region B, be expressed as client 1 and entered inside shop by shop doorway outdoors from shop, reach
Terminal 1-3, meets into shop logic judgment.
2) movement track two
The starting point 2-1 and terminal 2-3 of the movement track two of client 2 are respectively positioned on region A, and one of intermediate point 2-2
It in region B, is expressed as client 2 and passes by shop doorway outdoors from shop being back to shop again outdoors, and do not enter inside shop, no
It constitutes into shop behavior, is unsatisfactory for into shop logic judgment.
3) movement track three
The movement track three of client 3 is similar to the movement track two of client 2, and only terminal 2-3 is located at the area of the two sides rectangle b
Domain, as previously mentioned, this region is not belonging to current shop, client 3 is final and does not enter the region C represented inside shop, therefore
Movement track three is also unsatisfactory for into shop logic judgment.
Similar, the method for the shop logic judgment out includes:
By specified region division at shop region, shop doorway region, shop interior zone outdoors.
Action trail is analyzed, the action trail for meeting the following conditions simultaneously is judged to meeting shop logic judgment standard out:
1) starting point of action trail is within the scope of the interior zone of shop, 2) terminal of action trail is in shop outdoors regional scope,
3) more at least it is present in action trail in the regional scope of shop doorway, is patrolled otherwise, it is determined that action trail is unsatisfactory for out shop
Collect judgement.
On the basis of preceding method, the present invention further mentions a kind of passenger flow statistical system based on head and shoulder detection, the system
System integrates module, disengaging shop behavior including video acquisition device, video parsing module, personnel identity determination module, action trail
Determination module, counting module, label setup module.
Video acquisition device, to acquire the video image in a specified region, the specified region includes shop area outdoors
Domain, shop doorway region, shop interior zone.
Video parsing module, to extract the head and shoulder information of each personnel in video image, head and shoulder information includes at least head
Shoulder tracks ID, head and shoulder coordinate, and head and shoulder tracking ID is stored to a monitoring list.
Whether personnel identity determination module is salesman to analyze the corresponding personnel of each head and shoulder tracking ID.
Action trail integrates module, the head and shoulder coordinate of each head and shoulder tracking ID is integrated into action trail.
Disengaging shop behavior determination module, to determine whether the action trail of personnel constitutes the behavior of disengaging shop.
Counting module, to count passenger flow counting.
Label setup module analyzes institute shop interior zone is divided into several sub-regions according to item property
The movement track for having left the personnel of non-salesman in specified region is stated, the personnel is counted in the residence time of each sub-regions, presses
Corresponding consumption habit label is arranged as the personnel in the rule set according to one.
Further, the video parsing module includes video decoding sub-module, head and shoulder tests and analyzes submodule, face is examined
Survey analysis submodule, face characteristic value extraction and analysis submodule.
Video decoding sub-module, to by the decoding video stream of the video image in the specified region at several video frames
Module.
Head and shoulder tests and analyzes submodule, extracted corresponding to head and shoulder tracking ID, each head and shoulder tracking ID from video frame
Head and shoulder coordinate module.
Face datection analyzes submodule, to extract face coordinate corresponding to each head and shoulder tracking ID from video frame
Module.
Face characteristic value extraction and analysis submodule, to extract face corresponding to each head and shoulder tracking ID from video frame
The module of characteristic value.
To which the present invention is referred to based on the passenger flow statistical method and system of head and shoulder detection, is judged using head and shoulder detection technique
The disengaging shop behavior of personnel excludes salesman using face recognition technology, realizes the accurate passenger flow statistics in shop.
Although the present invention has been disclosed as a preferred embodiment, however, it is not to limit the invention.Skill belonging to the present invention
Has usually intellectual in art field, without departing from the spirit and scope of the present invention, when can be used for a variety of modifications and variations.Cause
This, the scope of protection of the present invention is defined by those of the claims.
Claims (10)
1. a kind of passenger flow statistical method based on head and shoulder detection, which is characterized in that the described method comprises the following steps:
S1: the video image in one specified region of acquisition extracts the head and shoulder information of each personnel in video image, head and shoulder information is at least
Including head and shoulder track ID, head and shoulder coordinate, by head and shoulder tracking ID store to one monitoring list, the specified region include shop outdoors
Region, shop doorway region, shop interior zone;
S2: successively analyzing each head and shoulder tracking ID, record the corresponding all head and shoulder coordinates of non-salesman user's head and shoulder tracking ID until
Non- salesman user leaves the specified region, to obtain the movement track of non-salesman user;
S3: whether the movement track that the personnel of the non-salesman in specified region have been left in judgement constitutes the behavior of disengaging shop, if so, objective
Flow accounting adds 1, if not, passenger flow counting adds 0;
S4: stopping tracking and the head and shoulder tracking ID is deleted from monitoring list;
S5: shop interior zone is divided by several sub-regions according to item property, analysis is described to have left specified region
The movement track of the personnel of non-salesman counts the personnel in the residence time of each sub-regions, and the rule set according to one is is somebody's turn to do
Corresponding consumption habit label is arranged in personnel.
2. the passenger flow statistical method according to claim 1 based on head and shoulder detection, which is characterized in that in step 2, successively divide
Analyse whether the corresponding personnel of each head and shoulder tracking ID are salesman, if so, by the identity of head and shoulder tracking ID in monitoring list
Attribute is labeled as salesman until the personnel corresponding to it leave the specified region, if not, by the head and shoulder in monitoring list
The identity attribute for tracking ID is labeled as non-salesman, while recording all head and shoulder coordinates of head and shoulder tracking ID until corresponding to it
Personnel leave the specified region, to obtain the movement track of the personnel.
3. the passenger flow statistical method according to claim 2 based on head and shoulder detection, which is characterized in that the every head of analysis
Whether the corresponding personnel of shoulder tracking ID are that the method for salesman includes:
It obtains head and shoulder and tracks ID, it is compared with the head and shoulder tracking ID in monitoring list to obtain its identity attribute, and
If head and shoulder tracking ID is not present in monitoring list or is present in monitoring list but do not do identity attribute label, obtain
It takes and the head and shoulder is tracked into face characteristic value corresponding to ID, the face characteristic value that will acquire is done with salesman's face characteristic value database
It compares, to judge whether it is salesman, if so, otherwise supervising head and shoulder tracking ID labeled as salesman in monitoring list
Depending on head and shoulder tracking ID is labeled as non-salesman in list.
4. the passenger flow statistical method according to claim 1 based on head and shoulder detection, which is characterized in that the method is also wrapped
It includes:
By the decoding video stream of the video image in the specified region at several video frames;
Head and shoulder detection and analysis, Face datection analysis, face characteristic value extraction and analysis are executed to each video frame, to obtain video frame
In each personnel head and shoulder track ID, head and shoulder coordinate, face coordinate, face characteristic value information;
The head and shoulder tracking ID that will acquire is stored to monitoring list, establish or update corresponding with its head and shoulder coordinate of head and shoulder tracking ID,
The mapping relations of face coordinate, face characteristic value information.
5. the passenger flow statistical method according to claim 4 based on head and shoulder detection, which is characterized in that the head that will acquire
Shoulder tracking ID is stored to monitoring list, while establising or updating head and shoulder tracking ID and corresponding head and shoulder coordinate, face coordinate, face
The mapping relations of characteristic value information refer to,
The head and shoulder tracking ID that will acquire compares with head and shoulder tracking ID existing in monitoring list, and 1) if head and shoulder tracking ID is not
It is present in monitoring list, is stored to monitoring list, establish corresponding head and shoulder coordinate database, face coordinate data
Library, face characteristic value information database store head and shoulder coordinate, face coordinate, face characteristic value information respectively to head and shoulder coordinate
Database, face coordinate database, in face characteristic value information database, 2) if head and shoulder tracking ID is present in monitoring list
In, head and shoulder coordinate, face coordinate, face characteristic value information are stored respectively to head and shoulder coordinate corresponding to head and shoulder tracking ID
Database, face coordinate database, in face characteristic value information database.
6. according to claim 1 to the passenger flow statistical method based on head and shoulder detection described in 5 any one, which is characterized in that
The movement track method that whether constitutes the behavior of disengaging shop of the personnel of the non-salesman for judging to have left specified region includes:
Action trail is taken turns doing into shop logic judgment and shop logic judgment out, will meet simultaneously into shop logic judgment standard and go out
The action trail of shop logic judgment standard is demarcated as constituting the behavior of disengaging shop.
7. the passenger flow statistical method according to claim 6 based on head and shoulder detection, which is characterized in that described to sentence into shop logic
Disconnected method includes:
By specified region division at shop region, shop doorway region, shop interior zone outdoors;
Action trail is analyzed, the action trail for meeting the following conditions simultaneously is judged to meeting into shop logic judgment standard: 1) row
For track starting point in shop outdoors regional scope, 2) terminal of action trail is within the scope of the interior zone of shop, 3) row
To be more at least present in the regional scope of shop doorway in track, otherwise, it is determined that action trail is unsatisfactory for sentencing into shop logic
It is disconnected.
8. the passenger flow statistical method according to claim 6 based on head and shoulder detection, which is characterized in that the shop logic out is sentenced
Disconnected method includes:
By specified region division at shop region, shop doorway region, shop interior zone outdoors;
Action trail is analyzed, the action trail for meeting the following conditions simultaneously is judged to meeting shop logic judgment standard out: 1) row
For track starting point within the scope of the interior zone of shop, 2) terminal of action trail is in shop outdoors regional scope, 3) row
To be more at least present in the regional scope of shop doorway in track, sentence otherwise, it is determined that action trail is unsatisfactory for out shop logic
It is disconnected.
9. the passenger flow statistical method according to claim 1 based on head and shoulder detection, which is characterized in that the method is also wrapped
It includes:
Shop interior zone is divided into several sub-regions according to item property, the non-shop in specified region has each been left in analysis
The movement track of the personnel of member is calculated what the personnel of each non-salesman for having left specified region stayed in each sub-regions
Time;
The total time t that the personnel of all non-salesmans in each subregion within the scope of a setting time stay is counted, is set in conjunction with same
In range of fixing time in each subregion commodity sale number h, each subregion commodity are calculated at this according to following formula
Fast-selling degree S within the scope of one setting time:
S=ω1×t+ω2×h
Wherein, ω1、ω2The weight factor of respectively total time t and sale number h, ω1+ω2=1.
10. a kind of passenger flow statistical system based on head and shoulder detection, which is characterized in that the system comprises:
Video acquisition device, to acquire the video image in a specified region, the specified region includes shop region, shop outdoors
Spread doorway region, shop interior zone;
Video parsing module, to extract the head and shoulder information of each personnel in video image, head and shoulder information is chased after including at least head and shoulder
Track ID, head and shoulder coordinate store head and shoulder tracking ID to a monitoring list;
Whether personnel identity determination module is salesman to analyze the corresponding personnel of each head and shoulder tracking ID;
Action trail integrates module, the head and shoulder coordinate of each head and shoulder tracking ID is integrated into action trail;
Disengaging shop behavior determination module, to determine whether the action trail of personnel constitutes the behavior of disengaging shop;
Counting module, to count passenger flow counting;
Label setup module, shop interior zone is divided into several sub-regions according to item property, analysis is described
The movement track for leaving the personnel of the non-salesman in specified region counts the personnel in the residence time of each sub-regions, according to one
Corresponding consumption habit label is arranged as the personnel in the rule set.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811367834.0A CN109448026A (en) | 2018-11-16 | 2018-11-16 | Passenger flow statistical method and system based on head and shoulder detection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811367834.0A CN109448026A (en) | 2018-11-16 | 2018-11-16 | Passenger flow statistical method and system based on head and shoulder detection |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109448026A true CN109448026A (en) | 2019-03-08 |
Family
ID=65553888
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811367834.0A Pending CN109448026A (en) | 2018-11-16 | 2018-11-16 | Passenger flow statistical method and system based on head and shoulder detection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109448026A (en) |
Cited By (28)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110310164A (en) * | 2019-07-30 | 2019-10-08 | 广州云从信息科技有限公司 | Image processing system, method, platform, machine readable media and equipment |
CN110399835A (en) * | 2019-07-26 | 2019-11-01 | 北京文安智能技术股份有限公司 | A kind of analysis method of personnel's residence time, apparatus and system |
CN110543868A (en) * | 2019-09-09 | 2019-12-06 | 福建省趋普物联科技有限公司 | Monitoring method and system based on face recognition and head and shoulder detection |
CN110659588A (en) * | 2019-09-02 | 2020-01-07 | 平安科技(深圳)有限公司 | Passenger flow volume statistical method and device and computer readable storage medium |
CN110659578A (en) * | 2019-08-26 | 2020-01-07 | 中国电子科技集团公司电子科学研究院 | Passenger flow statistical method, system and equipment based on detection and tracking technology |
CN110717885A (en) * | 2019-09-02 | 2020-01-21 | 平安科技(深圳)有限公司 | Customer number counting method and device, electronic equipment and readable storage medium |
CN110781748A (en) * | 2019-09-24 | 2020-02-11 | 上海商米科技集团股份有限公司 | IPC-based image processing method and camera |
CN110874589A (en) * | 2020-01-17 | 2020-03-10 | 南京甄视智能科技有限公司 | Face photo obtaining method and system |
CN110889339A (en) * | 2019-11-12 | 2020-03-17 | 南京甄视智能科技有限公司 | Head and shoulder detection-based dangerous area grading early warning method and system |
CN110930432A (en) * | 2019-11-19 | 2020-03-27 | 北京文安智能技术股份有限公司 | Video analysis method, device and system |
CN111104554A (en) * | 2019-12-23 | 2020-05-05 | 南京甄视智能科技有限公司 | Automatic door control method and system based on contour detection and head and shoulder detection |
CN111127096A (en) * | 2019-12-23 | 2020-05-08 | 南京甄视智能科技有限公司 | Commodity area passenger flow optimization statistical method and system based on head and shoulder and face recognition |
CN111738134A (en) * | 2020-06-18 | 2020-10-02 | 北京市商汤科技开发有限公司 | Method, device, equipment and medium for acquiring passenger flow data |
CN111768520A (en) * | 2020-06-04 | 2020-10-13 | 站坐(北京)科技有限公司 | Target detection device and method |
CN111832451A (en) * | 2020-07-01 | 2020-10-27 | 中国民用航空局信息中心 | Airworthiness monitoring process supervision system and method based on video data processing |
CN112070555A (en) * | 2020-09-16 | 2020-12-11 | 青岛维感科技有限公司 | Passenger flow monitoring method, device, system, channel and storage medium |
CN112351389A (en) * | 2019-08-09 | 2021-02-09 | 上海红星美凯龙悦家互联网科技有限公司 | Tracking method, device, terminal, medium and system for person track in commercial place |
CN112633204A (en) * | 2020-12-29 | 2021-04-09 | 厦门瑞为信息技术有限公司 | Accurate passenger flow statistical method, device, equipment and medium |
CN112668525A (en) * | 2020-12-31 | 2021-04-16 | 深圳云天励飞技术股份有限公司 | People flow counting method and device, electronic equipment and storage medium |
CN112818922A (en) * | 2021-02-25 | 2021-05-18 | 上海数川数据科技有限公司 | Store clerk identification method based on image |
CN112906439A (en) * | 2019-12-04 | 2021-06-04 | 上海稻知信息科技有限公司 | Passenger flow analysis method and system based on target tracking and behavior detection |
CN113095209A (en) * | 2021-04-07 | 2021-07-09 | 深圳海智创科技有限公司 | Crowd identification method and system for passenger flow and electronic equipment |
CN113160575A (en) * | 2021-03-15 | 2021-07-23 | 超级视线科技有限公司 | Traffic violation detection method and system for non-motor vehicles and drivers |
CN113221764A (en) * | 2021-05-18 | 2021-08-06 | 安徽工程大学 | Rapid pedestrian re-identification method |
CN113506132A (en) * | 2021-07-06 | 2021-10-15 | 树蛙信息科技(南京)有限公司 | Method and device for determining offline attention degree |
CN113610890A (en) * | 2021-07-02 | 2021-11-05 | 河南橡树智能科技有限公司 | Real-time personnel track two-dimensional plane display method based on multiple cameras |
US20230033062A1 (en) * | 2021-07-30 | 2023-02-02 | Fujitsu Limited | Non-transitory computer-readable recording medium, customer service detection method, and information processing device |
CN118015663A (en) * | 2024-04-09 | 2024-05-10 | 浙江深象智能科技有限公司 | Staff identification method, device and equipment |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090299814A1 (en) * | 2008-05-31 | 2009-12-03 | International Business Machines Corporation | Assessing personality and mood characteristics of a customer to enhance customer satisfaction and improve chances of a sale |
US8254633B1 (en) * | 2009-04-21 | 2012-08-28 | Videomining Corporation | Method and system for finding correspondence between face camera views and behavior camera views |
CN104217129A (en) * | 2014-09-22 | 2014-12-17 | 北京交通大学 | Passenger flow estimation method for urban rail road network |
CN104268578A (en) * | 2014-10-15 | 2015-01-07 | 深圳市晓舟科技有限公司 | Target recognition method implemented by combining small image comparison and fuzzy recognition |
CN105608419A (en) * | 2015-12-15 | 2016-05-25 | 上海微桥电子科技有限公司 | Passenger flow video detection and analysis system |
CN105844234A (en) * | 2016-03-21 | 2016-08-10 | 商汤集团有限公司 | People counting method and device based on head shoulder detection |
CN106355682A (en) * | 2015-07-08 | 2017-01-25 | 北京文安智能技术股份有限公司 | Video analysis method, device and system |
CN106897678A (en) * | 2017-02-08 | 2017-06-27 | 中国人民解放军军事医学科学院卫生装备研究所 | A kind of remote human face recognition methods of combination heartbeat signal, device and system |
CN107615298A (en) * | 2015-05-25 | 2018-01-19 | 彻可麦迪克私人投资有限公司 | Face identification method and system |
CN108051777A (en) * | 2017-12-01 | 2018-05-18 | 北京迈格威科技有限公司 | Method for tracing, device and the electronic equipment of target |
CN108564008A (en) * | 2018-03-28 | 2018-09-21 | 厦门瑞为信息技术有限公司 | A kind of real-time pedestrian and method for detecting human face based on ZYNQ |
CN108647242A (en) * | 2018-04-10 | 2018-10-12 | 北京天正聚合科技有限公司 | A kind of generation method and system of thermodynamic chart |
-
2018
- 2018-11-16 CN CN201811367834.0A patent/CN109448026A/en active Pending
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090299814A1 (en) * | 2008-05-31 | 2009-12-03 | International Business Machines Corporation | Assessing personality and mood characteristics of a customer to enhance customer satisfaction and improve chances of a sale |
US8254633B1 (en) * | 2009-04-21 | 2012-08-28 | Videomining Corporation | Method and system for finding correspondence between face camera views and behavior camera views |
CN104217129A (en) * | 2014-09-22 | 2014-12-17 | 北京交通大学 | Passenger flow estimation method for urban rail road network |
CN104268578A (en) * | 2014-10-15 | 2015-01-07 | 深圳市晓舟科技有限公司 | Target recognition method implemented by combining small image comparison and fuzzy recognition |
CN107615298A (en) * | 2015-05-25 | 2018-01-19 | 彻可麦迪克私人投资有限公司 | Face identification method and system |
CN106355682A (en) * | 2015-07-08 | 2017-01-25 | 北京文安智能技术股份有限公司 | Video analysis method, device and system |
CN105608419A (en) * | 2015-12-15 | 2016-05-25 | 上海微桥电子科技有限公司 | Passenger flow video detection and analysis system |
CN105844234A (en) * | 2016-03-21 | 2016-08-10 | 商汤集团有限公司 | People counting method and device based on head shoulder detection |
CN106897678A (en) * | 2017-02-08 | 2017-06-27 | 中国人民解放军军事医学科学院卫生装备研究所 | A kind of remote human face recognition methods of combination heartbeat signal, device and system |
CN108051777A (en) * | 2017-12-01 | 2018-05-18 | 北京迈格威科技有限公司 | Method for tracing, device and the electronic equipment of target |
CN108564008A (en) * | 2018-03-28 | 2018-09-21 | 厦门瑞为信息技术有限公司 | A kind of real-time pedestrian and method for detecting human face based on ZYNQ |
CN108647242A (en) * | 2018-04-10 | 2018-10-12 | 北京天正聚合科技有限公司 | A kind of generation method and system of thermodynamic chart |
Cited By (35)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110399835A (en) * | 2019-07-26 | 2019-11-01 | 北京文安智能技术股份有限公司 | A kind of analysis method of personnel's residence time, apparatus and system |
CN110399835B (en) * | 2019-07-26 | 2024-04-02 | 北京文安智能技术股份有限公司 | Analysis method, device and system for personnel residence time |
CN110310164A (en) * | 2019-07-30 | 2019-10-08 | 广州云从信息科技有限公司 | Image processing system, method, platform, machine readable media and equipment |
CN112351389A (en) * | 2019-08-09 | 2021-02-09 | 上海红星美凯龙悦家互联网科技有限公司 | Tracking method, device, terminal, medium and system for person track in commercial place |
CN110659578A (en) * | 2019-08-26 | 2020-01-07 | 中国电子科技集团公司电子科学研究院 | Passenger flow statistical method, system and equipment based on detection and tracking technology |
CN110659588A (en) * | 2019-09-02 | 2020-01-07 | 平安科技(深圳)有限公司 | Passenger flow volume statistical method and device and computer readable storage medium |
CN110717885A (en) * | 2019-09-02 | 2020-01-21 | 平安科技(深圳)有限公司 | Customer number counting method and device, electronic equipment and readable storage medium |
CN110543868A (en) * | 2019-09-09 | 2019-12-06 | 福建省趋普物联科技有限公司 | Monitoring method and system based on face recognition and head and shoulder detection |
CN110781748A (en) * | 2019-09-24 | 2020-02-11 | 上海商米科技集团股份有限公司 | IPC-based image processing method and camera |
CN110889339A (en) * | 2019-11-12 | 2020-03-17 | 南京甄视智能科技有限公司 | Head and shoulder detection-based dangerous area grading early warning method and system |
CN110930432A (en) * | 2019-11-19 | 2020-03-27 | 北京文安智能技术股份有限公司 | Video analysis method, device and system |
CN112906439A (en) * | 2019-12-04 | 2021-06-04 | 上海稻知信息科技有限公司 | Passenger flow analysis method and system based on target tracking and behavior detection |
CN111127096B (en) * | 2019-12-23 | 2022-08-16 | 南京甄视智能科技有限公司 | Commodity area passenger flow optimization statistical method and system based on head and shoulder and face recognition |
CN111104554A (en) * | 2019-12-23 | 2020-05-05 | 南京甄视智能科技有限公司 | Automatic door control method and system based on contour detection and head and shoulder detection |
CN111127096A (en) * | 2019-12-23 | 2020-05-08 | 南京甄视智能科技有限公司 | Commodity area passenger flow optimization statistical method and system based on head and shoulder and face recognition |
CN111104554B (en) * | 2019-12-23 | 2022-08-16 | 南京甄视智能科技有限公司 | Automatic door control method and system based on contour detection and head and shoulder detection |
CN110874589A (en) * | 2020-01-17 | 2020-03-10 | 南京甄视智能科技有限公司 | Face photo obtaining method and system |
CN111768520A (en) * | 2020-06-04 | 2020-10-13 | 站坐(北京)科技有限公司 | Target detection device and method |
CN111738134A (en) * | 2020-06-18 | 2020-10-02 | 北京市商汤科技开发有限公司 | Method, device, equipment and medium for acquiring passenger flow data |
CN111832451A (en) * | 2020-07-01 | 2020-10-27 | 中国民用航空局信息中心 | Airworthiness monitoring process supervision system and method based on video data processing |
CN112070555A (en) * | 2020-09-16 | 2020-12-11 | 青岛维感科技有限公司 | Passenger flow monitoring method, device, system, channel and storage medium |
CN112633204A (en) * | 2020-12-29 | 2021-04-09 | 厦门瑞为信息技术有限公司 | Accurate passenger flow statistical method, device, equipment and medium |
CN112668525A (en) * | 2020-12-31 | 2021-04-16 | 深圳云天励飞技术股份有限公司 | People flow counting method and device, electronic equipment and storage medium |
CN112668525B (en) * | 2020-12-31 | 2024-05-07 | 深圳云天励飞技术股份有限公司 | People flow counting method and device, electronic equipment and storage medium |
CN112818922B (en) * | 2021-02-25 | 2022-08-02 | 上海数川数据科技有限公司 | Shop assistant identification method based on image |
CN112818922A (en) * | 2021-02-25 | 2021-05-18 | 上海数川数据科技有限公司 | Store clerk identification method based on image |
CN113160575A (en) * | 2021-03-15 | 2021-07-23 | 超级视线科技有限公司 | Traffic violation detection method and system for non-motor vehicles and drivers |
CN113095209B (en) * | 2021-04-07 | 2024-05-31 | 深圳海智创科技有限公司 | Crowd identification method and system for passenger flow and electronic equipment |
CN113095209A (en) * | 2021-04-07 | 2021-07-09 | 深圳海智创科技有限公司 | Crowd identification method and system for passenger flow and electronic equipment |
CN113221764B (en) * | 2021-05-18 | 2023-04-28 | 安徽工程大学 | Rapid pedestrian re-identification method |
CN113221764A (en) * | 2021-05-18 | 2021-08-06 | 安徽工程大学 | Rapid pedestrian re-identification method |
CN113610890A (en) * | 2021-07-02 | 2021-11-05 | 河南橡树智能科技有限公司 | Real-time personnel track two-dimensional plane display method based on multiple cameras |
CN113506132A (en) * | 2021-07-06 | 2021-10-15 | 树蛙信息科技(南京)有限公司 | Method and device for determining offline attention degree |
US20230033062A1 (en) * | 2021-07-30 | 2023-02-02 | Fujitsu Limited | Non-transitory computer-readable recording medium, customer service detection method, and information processing device |
CN118015663A (en) * | 2024-04-09 | 2024-05-10 | 浙江深象智能科技有限公司 | Staff identification method, device and equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109448026A (en) | Passenger flow statistical method and system based on head and shoulder detection | |
US20170169297A1 (en) | Computer-vision-based group identification | |
US10943204B2 (en) | Realtime video monitoring applied to reduce customer wait times | |
US8665333B1 (en) | Method and system for optimizing the observation and annotation of complex human behavior from video sources | |
CN109784162A (en) | A kind of identification of pedestrian behavior and trace tracking method | |
CN109359625A (en) | The method and system of customer identification is judged based on head and shoulder detection and face recognition technology | |
JP3800257B2 (en) | Attention information measurement method and apparatus, and various systems using the same | |
Liciotti et al. | Person re-identification dataset with rgb-d camera in a top-view configuration | |
CN109272347A (en) | A kind of statistical analysis technique and system of shops's volume of the flow of passengers | |
JP4069932B2 (en) | Human detection device and human detection method | |
US9245247B2 (en) | Queue analysis | |
US20090217315A1 (en) | Method and system for audience measurement and targeting media | |
US20080215462A1 (en) | Still image shopping event monitoring and analysis system and method | |
CN105608419A (en) | Passenger flow video detection and analysis system | |
JPH1049718A (en) | Device for counting number of passing person and system for managing number of visiting and leaning person using the device | |
CN109344765A (en) | A kind of intelligent analysis method entering shop personnel analysis for chain shops | |
CN111814510B (en) | Method and device for detecting legacy host | |
CA2448452A1 (en) | Automatic classification and/or counting system | |
CN110889719A (en) | Advertisement pushing system and method based on user consumption characteristics | |
Okamoto et al. | Classification of Pedestrian Behavior in a Shopping Mall based on LRF and Camera Observations. | |
CN111476600B (en) | Statistical analysis method for audience numbers of direct ladder advertisement | |
CN112347907A (en) | 4S store visitor behavior analysis system based on Reid and face recognition technology | |
Herviana et al. | The prototype of in-store visitor and people passing counters using single shot detector performed by OpenCV | |
Popa et al. | Detecting customers’ buying events on a real-life database | |
CN109325800A (en) | A kind of working method of supermarket's intelligent commodity shelf based on computer vision |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190308 |