CN108501954A - A kind of gesture identification method, device, automobile and storage medium - Google Patents
A kind of gesture identification method, device, automobile and storage medium Download PDFInfo
- Publication number
- CN108501954A CN108501954A CN201810288399.6A CN201810288399A CN108501954A CN 108501954 A CN108501954 A CN 108501954A CN 201810288399 A CN201810288399 A CN 201810288399A CN 108501954 A CN108501954 A CN 108501954A
- Authority
- CN
- China
- Prior art keywords
- barrier
- gesture
- people
- point cloud
- profile
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/107—Static hand or arm
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the invention discloses a kind of gesture identification method, device, automobile and storage mediums.Wherein, method includes：Obtain surrounding obstacle object point cloud in vehicle traveling process；Each barrier profile is determined according to the obstacle object point cloud of surrounding；Judge whether each barrier is people respectively according to the characteristic of each barrier profile；If barrier is people, its gesture is identified according to advance trained traffic police's gesture model.The embodiment of the present invention obtains each barrier profile by the barrier point cloud data for obtaining and analyzing around in vehicle traveling process, outline data is input in advance trained machine learning model, judge whether each barrier is people according to the output result of model, to determining its gesture is further identified for the barrier of people, solve the problems, such as that intelligent automobile cannot fully cognitive disorders object information and react in the process of moving, so that intelligent automobile is can receive the instruction that extraneous people conveys in the process of moving, and corresponding reaction is made according to instruction.
Description
Technical field
The present embodiments relate to mode identification technology more particularly to a kind of gesture identification method, device, automobile and storages
Medium.
Background technology
Intelligent vehicle is a comprehensive system for integrating the functions such as environment sensing, programmed decision-making, multi-grade auxiliary driving
System, it, which is concentrated, has used computer, modern sensing, information fusion, communication, artificial intelligence and the technologies such as has automatically controlled.Intelligent vapour
Vehicle in the process of moving, is mostly identified to grasp traffic information barrier and curb information by laser sensor, it is ensured that
The safety of intelligent automobile.
But existing intelligent automobile is not enough obstacle recognition by laser sensor, not to barrier
Category attribute makes differentiation, and differentiated reaction cannot be made according to the difference of barrier.In particular, intelligent automobile is in driving process
In be unable to real-time reception people from external world reception and registration order, such as traffic police gesture, lack flexible property.
Invention content
A kind of gesture identification method of offer of the embodiment of the present invention, device, automobile and storage medium, to realize that intelligent automobile exists
It can recognize that whether the attribute of barrier is people during traveling, and identify the instruction that people is conveyed, make corresponding reaction.
In a first aspect, an embodiment of the present invention provides a kind of gesture identification method, this method includes：
Obtain surrounding obstacle object point cloud in vehicle traveling process；
Each barrier profile is determined according to the obstacle object point cloud around described；
Judge whether each barrier is people respectively according to the characteristic of each barrier profile；
If barrier is people, its gesture is identified according to advance trained traffic police's gesture model.
Second aspect, the embodiment of the present invention additionally provide a kind of gesture identifying device, which includes：
Data acquisition module, for obtaining surrounding obstacle object point cloud in vehicle traveling process；
Profile extraction module, for determining each barrier profile according to the obstacle object point cloud of the surrounding；
Barrier classification judgment module, for judging each barrier respectively according to the characteristic of each barrier profile
Whether it is people；
Gesture recognition module, for when barrier is people, its hand to be identified according to advance trained traffic police's gesture model
Gesture.
The third aspect, the embodiment of the present invention additionally provide a kind of automobile, which includes：
Laser sensor for obtaining the obstacle object point cloud around in the vehicle traveling process, and obtains each barrier
Distance；
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 processing
Device realizes the gesture identification method as described in any in the embodiment of the present invention.
Fourth aspect, the embodiment of the present invention additionally provide a kind of computer readable storage medium, are stored thereon with computer
Program realizes the gesture identification method as described in any in the embodiment of the present invention when program is executed by processor.
The embodiment of the present invention is obtained respectively by the barrier point cloud data for obtaining and analyzing around in vehicle traveling process
Each barrier outline data is input in advance trained machine learning model, according to machine learning mould by barrier profile
The output result of block judges whether each barrier is people, further identifies its gesture for the barrier of people to determining, solves
Intelligent automobile in the process of moving cannot fully cognitive disorders object information and problem of reacting, reached intelligent automobile and travelled
It can receive the instruction that extraneous people conveys in the process, and the effect reacted accordingly made according to instruction.
Description of the drawings
Fig. 1 is the flow chart of the gesture identification method in the embodiment of the present invention one；
Fig. 2 is the flow chart of the gesture identification method in the embodiment of the present invention two；
Fig. 3 is the structural schematic diagram of the gesture identifying device in the embodiment of the present invention three；
Fig. 4 is the illustrative view of functional configuration of the automobile in the embodiment of the present invention four.
Specific implementation mode
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limitation of the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
Fig. 1 is the flow chart for the gesture identification method that the embodiment of the present invention one provides, and the present embodiment is applicable to intelligent vapour
Vehicle identifies the case where external information in the process of moving.Specifically, intelligent vehicle is exactly to be increased on the basis of common vehicle
The devices such as advanced sensor (radar, camera shooting), controller, actuator, by vehicle-mounted sensor-based system and information terminal realize with
The intelligent information on people, vehicle, road etc. exchanges, and vehicle is made to have intelligent environment sensing ability, can automatically analyze vehicle traveling
Safety and precarious position, and vehicle is made to be arrived at according to the wish of people, it is final to realize the purpose for substituting people to operate.In intelligence
Can be in vehicle traveling process, gesture identification method can be executed by gesture identifying device, the device may be used software and/or
The mode of hardware is realized.
As shown in Figure 1, this method specifically includes：
S110, surrounding obstacle object point cloud in vehicle traveling process is obtained.
Wherein, obstacle object point cloud refers to the point data set of barrier appearance surfaces, also referred to as puts cloud.In the present embodiment
In be the barrier point cloud data obtained by the laser sensor on intelligent automobile, use three-dimensional laser scanner
Or the obtained point cloud point quantity of photographic-type scanner is bigger and than comparatively dense, is usually point off density cloud, data volume is bigger, institute
The obstacle information that can be obtained is also abundanter.
Specifically, automobile is in the process of moving, surrounding barrier may include to be travelled on track beside it
Automobile, pedestrian, curb, greenbelt, great Shu, traffic post etc..Wherein, the automobile that is travelled on the track of side, pedestrian, curb information can
The automobile moment is reminded to pay attention to spacing, certain safe distance with the holdings such as the automobile of form, pedestrian, curb on the track of side.It is logical
Often, the traffic post of running car direction of advance is provided with traffic lights and/or traffic police, to traveling automobile to indicate, keep straight on, turn
Curved or stopping, the automobile of traveling are braked according to the gesture of traffic police, so as to realize safety traffic.
S120, each barrier profile is determined according to the obstacle object point cloud around described.
In general, traffic post and/or traffic police in the direction of advance of running car, need to carry out acquired point cloud data
Extraction, identifies the profile information of each barrier.It can specifically be realized by following steps：
First, the obstacle object point cloud of the surrounding is filtered.It is stochastical sampling consistency used by the present embodiment
Method point cloud data is filtered, it can pass through iterative manner from one group of observation data set including noise
The parameter for estimating mathematical model, to filter out noise data.Wherein, weather conditions, the distance for obtaining point cloud data, sensor
Degree of stability all may be generate noise the reason of.
Then, the obstacle object point cloud for dividing the filtered surrounding, obtains respective cloud of each barrier.Specifically,
The method that region growing is used in the present embodiment is split point cloud data, obtains the point cloud number of each barrier one by one
According to.The basic thought of algorithm of region growing is to gather the point cloud data with similar quality to constitute a region.Specifically
When realization, first the region of segmentation look for a sub-pixel as grow starting point, then by sub-pixel surrounding neighbors with kind
Sub-pixel has the pixel (being determined according to certain pre-determined growth criterion) of same or similar property to be merged into sub-pixel institute
In the zone, then process above is repeated using these newly-increased pixels as new growing point.Illustratively, such as barrier
For traffic police, traffic police is to stand on the ground, and the point cloud data of traffic police's body is different from the road surface point cloud data property of its underfooting
, therefore can divide and be determined as different regions, belong to different barriers.
Finally, then it is respectively to put cloud using each barrier and be fitted respectively to obtain the profile of each barrier.Utilize least square
Method, fitting respectively obtains each contour line of each barrier, to obtain the profile of each barrier.
S130, judge whether each barrier is people respectively according to the characteristic of each barrier profile.
Wherein, the characteristic of each barrier profile is that can indicate the data of barrier profile characteristic, is from each barrier
Common denominator data can be extracted by hindering in object data, to be compared with humanoid masterplate.
If carried out object identification in the picture, it is related to whether there is object in specific region, passes through SVM
(Support Vector Machine, support vector machines) machine learning can be made positive negative sample in the case that sample size is less
Good discrimination.Therefore, by machine learning, relevant mode can be established in advance using the outline data of a large amount of people as learning sample
Type judged by the way that the outline data feature of each barrier to be input in model, finally determine the barrier whether be
People.
Preferably, before judging whether each barrier is people respectively according to the characteristic of each barrier profile,
The method further includes：The profile of each barrier is normalized.The spy of each barrier profile after normalization
Sign data are all compared under unified each standard with the contour feature data in the model pre-established, are obtained more accurate
Judging result.
If S140, barrier are people, its gesture is identified according to advance trained traffic police's gesture model.
Wherein, advance trained traffic police's gesture model, is by SVM machine learning methods, by ten kinds of common traffic police
Gesture is learnt as sample, obtains traffic police's gesture model, and barrier profile Hu squares are then input to traffic police's gesture
In model, and then the posture that can obtain the barrier belongs to the probability of ten kinds of traffic police's gestures, determines the gesture of barrier.
Further, when being determined that barrier is people, and identify that the gesture of the barrier belongs to a certain gesture of traffic police,
Automobile will execute corresponding operation according to the gesture identified.To achieve the purpose that safety traffic.
The technical solution of the present embodiment, by obtaining and analyzing the barrier point cloud data around in vehicle traveling process
Each barrier profile is obtained, each barrier outline data is input in advance trained machine learning model, according to machine
The output result of study module judges whether each barrier is people, its gesture is further identified for the barrier of people to determining,
It solves the problems, such as that intelligent automobile cannot fully cognitive disorders object information and make a response in the process of moving, has reached intelligent automobile
It can receive the instruction that extraneous people conveys in the process of moving, and the effect reacted accordingly made according to instruction.
Embodiment two
Fig. 2 is the flow chart of gesture identification method provided by Embodiment 2 of the present invention, and the present embodiment is in the various embodiments described above
On the basis of advanced optimize.As shown in Fig. 2, this method specifically includes：
S210, surrounding obstacle object point cloud in vehicle traveling process is obtained.
Preferably, in obtaining vehicle traveling process before surrounding obstacle object point cloud, this method further includes：It is true respectively
Determine distance of the laser sensor to each barrier peak and the horizontal distance to each barrier；According to laser sensor to each barrier
Hinder the distance of object peak and determines the height of each barrier to the horizontal distance of each barrier.
Correspondingly, surrounding obstacle object point cloud includes in acquisition vehicle traveling process：Obtain its in vehicle traveling process
The height of surrounding meets the obstacle object point cloud of preset condition.
Specifically, the purpose in the present embodiment is to identify the gesture information of traffic police, the height interval range based on adult,
An obstacle height section is pre-set, can be illustratively 1 meter -4 meters, not in this interval range, can be sentenced
The disconnected barrier is not people, it may be possible to the barriers such as big tree, building or bushes, then such barrier can not be obtained
Point cloud data improves the efficiency of gesture identification to mitigate the burden of memory space and data processing unit.
S220, each barrier profile is determined according to the obstacle object point cloud around described.
S230, the Hu squares of each barrier profile are extracted as characteristic；
In the present embodiment using the Hu squares of each barrier profile as the characteristic of each barrier.Since Hu squares are by 7
A not bending moment constitutes one group of characteristic quantity, and there is rotation, zooming and panning invariance to be suitable for image area characteristics and describe.
S240, the Hu squares of each barrier profile are input in advance trained humanoid model whether obtain each barrier
For the first probability of people.
Wherein, trained humanoid model is obtained by the outline data to a large amount of people carries out machine learning in advance
, after the characteristic of any barrier profile is input in the model, you can obtaining inputted barrier profile is
No the first probability for people.
S250, compare the first probability and the first predetermined threshold value, judge whether each barrier is people according to comparison result.
Wherein the first predetermined threshold value is determined according to the accuracy rate of advance trained humanoid model judging result.Specifically
, when the first probability that a barrier is people is more than the first predetermined threshold value, it may be determined that the barrier is behaved.Sentenced according to this
The order of accuarcy for determining result determines the first predetermined threshold value.
S260, it will determine that the Hu squares of the barrier profile for people are input in traffic police's gesture model that training obtains in advance,
Determine that the barrier profile belongs to the second probability for making each traffic police's gesture respectively.
Wherein, advance trained traffic police's gesture model is by SVM machine learning methods, illustratively, often by ten kinds
The traffic police's gesture used learns as sample, obtains traffic police's gesture model, is then input to barrier profile Hu squares
In traffic police's gesture model, and then the posture that can obtain the barrier belongs to the probability of ten kinds of traffic police's gestures.So, in the second probability
Then contain the probability data corresponding to each traffic police's gesture.
S270, each second probability of comparison and the second predetermined threshold value, traffic police's gesture is determined according to comparison result.
Specifically, by taking traffic police's gesture model that advance training obtains includes ten common traffic police's gestures as an example, will appoint
The contour feature data of one barrier are input in traffic police's gesture model, can be obtained corresponding to each common traffic police's gesture
To second probability, then, will be greater than in the second probability of the second predetermined threshold value corresponding to maximum second probability of probability value
Traffic police's gesture, be determined as the gesture that the barrier is made.Further, if the barrier corresponds to each traffic police's gesture
The second probability be respectively less than the second predetermined threshold value, then it was determined that the barrier does not make any gesture.Wherein, second is default
Threshold value determination method is the same as the first predetermined threshold value.
The technical solution of the present embodiment, by obtaining and analyzing the barrier point cloud data around in vehicle traveling process
Each barrier profile is obtained, each barrier outline data is input in advance trained machine learning model, according to machine
The output result of study module judges whether each barrier is people, its gesture is further identified for the barrier of people to determining,
It solves the problems, such as that intelligent automobile cannot fully cognitive disorders object and react in the process of moving, has reached intelligent automobile and be expert at
It can receive the instruction that extraneous people conveys during sailing, and the effect reacted accordingly made according to instruction.In addition, obtaining
Barrier is screened before barrier point cloud data, the burden of memory space and data processing unit can be mitigated, is improved
The efficiency of gesture identification.Using Hu squares as characteristic, keeps the comparison result of sample and model with uniformity, improve hand
The accuracy rate of gesture identification.
Embodiment three
Fig. 3 show the structural schematic diagram of the gesture identifying device of the offer of the embodiment of the present invention three, and the present embodiment is applicable
In intelligent automobile identifies external information in the process of moving the case where.As shown in figure 3, the gesture identifying device specifically includes：
Data acquisition module 310, profile extraction module 320, barrier classification judgment module 330 and gesture recognition module 340.
Wherein, data acquisition module 310, for obtaining surrounding obstacle object point cloud in vehicle traveling process；Profile carries
Modulus block 320, for determining each barrier profile according to the obstacle object point cloud of the surrounding；Barrier classification judgment module 330,
For judging whether each barrier is people respectively according to the characteristic of each barrier profile；Gesture recognition module 340 is used
In when barrier is people, its gesture is identified according to advance trained traffic police's gesture model.
The technical solution of the present embodiment, by obtaining and analyzing the barrier point cloud data around in vehicle traveling process
Each barrier profile is obtained, each barrier outline data is input in advance trained machine learning model, according to machine
The output result of study module judges whether each barrier is people, its gesture is further identified for the barrier of people to determining,
It solves the problems, such as that intelligent automobile cannot fully cognitive disorders object information and react in the process of moving, has reached intelligent automobile
It can receive the instruction that extraneous people conveys in the process of moving, and the effect reacted accordingly made according to instruction.
Further, profile extraction module 320 specifically includes：
Filter unit is filtered for the obstacle object point cloud to the surrounding；
Point cloud segmentation unit, the obstacle object point cloud for dividing the filtered surrounding, it is respective to obtain each barrier
Point cloud；
Contour fitting unit is fitted respectively for respectively being put cloud using each barrier and obtains the profile of each barrier.
Further, barrier classification judgment module 330 specifically includes：
Characteristic extraction unit extracts the Hu squares of each barrier profile as characteristic；
First Model Matching unit, for the Hu squares of each barrier profile to be input in advance trained humanoid model
Obtain each barrier whether be people the first probability；
First comparing unit judges each barrier for comparing the first probability and the first predetermined threshold value according to comparison result
Whether it is people.
Further, gesture recognition module 340 specifically includes：
Second Model Matching unit, for will determine that the Hu squares of the barrier profile for people are input to what training in advance obtained
In traffic police's gesture model, determine that the barrier profile belongs to the second probability for making each traffic police's gesture respectively；
Second comparing unit is used for more each second probability and the second predetermined threshold value, traffic police's hand is determined according to comparison result
Gesture.
Preferably, gesture identifying device further includes obstacle height acquisition module, for being crossed in the acquisition garage
In journey before surrounding obstacle object point cloud, camera laser sensor is determined to the distance of each barrier peak respectively and is arrived
The horizontal distance of each barrier；According to the camera laser sensor to the distance of each barrier peak and arrive each barrier
Horizontal distance determine the height of each barrier.
Correspondingly, data acquisition module 310 is additionally operable to：It obtains surrounding height in vehicle traveling process and meets default item
The obstacle object point cloud of part.
Further, gesture identifying device further includes brake module, for according to the gesture identified, executing corresponding behaviour
Make.
Further, gesture identifying device further includes data normalization module, for according to each barrier profile
Characteristic judge whether each barrier is people respectively before, the profile of each barrier is normalized.
Above-mentioned gesture identifying device can perform the gesture identification method that any embodiment of the present invention is provided, and have the side of execution
The corresponding function module of method and advantageous effect.
Example IV
Fig. 4 is the illustrative view of functional configuration of the automobile in the embodiment of the present invention four.Fig. 4 is shown suitable for being used for realizing this hair
The block diagram of the functional structure 412 of the exemplary automobile of bright embodiment.The functional structure chart 412 for the automobile that Fig. 4 is shown is only
One example should not bring any restrictions to the function and use scope of the embodiment of the present invention.
As shown in figure 4, the functional structure 412 of automobile is showed in the form of universal computing device.The functional structure 412 of automobile
Component can include but is not limited to：One or more processor or processing unit 416, one or more laser sensors
436, system storage 428, the bus 418 of connection different system component (including system storage 428 and processing unit 416).
Bus 418 indicates one or more in a few class bus structures, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using the arbitrary bus structures in a variety of bus structures.It lifts
For example, these architectures include but not limited to industry standard architecture (ISA) bus, microchannel architecture (MAC)
Bus, enhanced isa bus, Video Electronics Standards Association (VESA) local bus and peripheral component interconnection (PCI) bus.
The functional structure 412 of automobile typically comprises a variety of computer system readable media.These media can be any
The usable medium that can be accessed by the functional structure 412 of automobile, including volatile and non-volatile media, movably and can not
Mobile medium.
System storage 428 may include the computer system readable media of form of volatile memory, such as deposit at random
Access to memory (RAM) 430 and/or cache memory 432.The functional structure 412 of automobile may further include it is other can
Movement/immovable, volatile/non-volatile computer system storage medium.Only as an example, storage system 434 can be with
For reading and writing immovable, non-volatile magnetic media (Fig. 4 do not show, commonly referred to as " hard disk drive ").Although in Fig. 4 not
It shows, can provide for the disc driver to moving non-volatile magnetic disk (such as " floppy disk ") read-write, and to removable
The CD drive of dynamic anonvolatile optical disk (such as CD-ROM, DVD-ROM or other optical mediums) read-write.In these situations
Under, each driver can be connected by one or more data media interfaces with bus 418.Memory 428 may include
There is one group of (for example, at least one) program module, these program modules to be configured at least one program product, the program product
To execute the function of various embodiments of the present invention.
Laser sensor (one or more) is for obtaining in vehicle traveling process, the point cloud data of peripheral obstacle,
Then it transmits data to one or more processor or processing unit 416 carries out data processing, to carry out barrier
Identification, and determine the identification for the gesture for being people's barrier.
Program/utility 440 with one group of (at least one) program module 442, can be stored in such as memory
In 428, such program module 442 includes but not limited to operating system, one or more application program, other program modules
And program data, the realization of network environment may be included in each or certain combination in these examples.Program module 442
Usually execute the function and/or method in embodiment described in the invention.
Server 412 can also be with one or more external equipments 414 (such as keyboard, sensing equipment, display 424 etc.)
Communication, can also be enabled a user to one or more equipment interact with the server 412 communicate, and/or with make the vapour
Any equipment that the functional structure 412 of vehicle can be communicated with one or more of the other computing device (such as network interface card, modulation /demodulation
Device etc.) communication.This communication can be carried out by input/output (I/O) interface 422.Also, server 412 can also lead to
Cross network adapter 420 and one or more network (such as LAN (LAN), wide area network (WAN) and/or public network, example
Such as internet) communication.As shown, network adapter 420 is communicated by bus 418 with other modules of server 412.It should
Understand, although not shown in fig 4, other hardware and/or software module can be used in conjunction with the functional structure 412 of automobile, including
But it is not limited to：Microcode, device driver, redundant processing unit, external disk drive array, RAID system, tape drive
And data backup storage system etc..
Processing unit 416 is stored in program in system storage 428 by operation, to perform various functions using with
And data processing, such as realize the gesture identification method that the embodiment of the present invention is provided, this method includes mainly：
Obtain surrounding obstacle object point cloud in vehicle traveling process；
Each barrier profile is determined according to the obstacle object point cloud around described；
Judge whether each barrier is people respectively according to the characteristic of each barrier profile；
If barrier is people, its gesture is identified according to advance trained traffic police's gesture model.
Embodiment five
The embodiment of the present invention five additionally provides a kind of computer readable storage medium, is stored thereon with computer program, should
Realize that the gesture identification method provided such as the embodiment of the present invention, this method include mainly when program is executed by processor：
Obtain surrounding obstacle object point cloud in vehicle traveling process；
Each barrier profile is determined according to the obstacle object point cloud around described；
Judge whether each barrier is people respectively according to the characteristic of each barrier profile；
If barrier is people, its gesture is identified according to advance trained traffic police's gesture model.The embodiment of the present invention
The arbitrary combination of one or more computer-readable media may be used in computer storage media.Computer-readable medium can
To be computer-readable signal media or computer readable storage medium.Computer readable storage medium for example can be ---
But be not limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or arbitrary above group
It closes.The more specific example (non exhaustive list) of computer readable storage medium includes：Electricity with one or more conducting wires
Connection, portable computer diskette, hard disk, random access memory (RAM), read-only memory (ROM), erasable type are programmable only
Read memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic storage
Device or above-mentioned any appropriate combination.In this document, computer readable storage medium can any be included or deposit
The tangible medium of program is stored up, which can be commanded the either device use or in connection of execution system, device.
Computer-readable signal media may include in a base band or as the data-signal that a carrier wave part is propagated,
Wherein carry computer-readable program code.Diversified forms may be used in the data-signal of this propagation, 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 instruction execution system, device either device use or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
It can be write with one or more programming languages or combinations thereof for executing the computer that operates of the present invention
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
Further include conventional procedural programming language-such as " such as " language or similar programming language.Program code can be with
It fully executes, partly execute on the user computer on the user computer, being executed as an independent software package, portion
Divide and partly executes or executed on a remote computer or server completely on the remote computer on the user computer.
Be related in the situation of remote computer, remote computer can pass through the network of any kind --- including LAN (LAN) or
The domain wide area network (WAN) is connected to subscriber computer, or, it may be connected to outer computer (such as carried using Internet service
It is connected by internet for quotient).
Note that above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The present invention is not limited to specific embodiments described here, can carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out to the present invention by above example
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
May include other more equivalent embodiments, and the scope of the present invention is determined by scope of the appended claims.
Claims (10)
1. a kind of gesture identification method, which is characterized in that including：
Obtain surrounding obstacle object point cloud in vehicle traveling process；
Each barrier profile is determined according to the obstacle object point cloud around described；
Judge whether each barrier is people respectively according to the characteristic of each barrier profile；
If barrier is people, its gesture is identified according to advance trained traffic police's gesture model.
2. gesture identification method according to claim 1, which is characterized in that the obstacle object point cloud according to the surrounding
Determine that each barrier profile includes：
Obstacle object point cloud around described is filtered；
The obstacle object point cloud for dividing the filtered surrounding, obtains respective cloud of each barrier；
Cloud is respectively put using each barrier be fitted respectively obtain the profile of each barrier.
3. gesture identification method according to claim 1, which is characterized in that the spy according to each barrier profile
Sign data judge whether each barrier is that people includes respectively：
The Hu squares of each barrier profile are extracted as characteristic；
The Hu squares of each barrier profile are input to and obtain whether each barrier is the of people in advance trained humanoid model
One probability；
Compare the first probability and the first predetermined threshold value, judges whether each barrier is people according to comparison result.
4. gesture identification method according to claim 3, which is characterized in that the basis trained traffic police's gesture in advance
Its gesture of Model Identification includes：
It will determine that the Hu squares of the barrier profile for people are input in traffic police's gesture model that training obtains in advance, determining respectively should
Barrier profile belongs to the second probability for making each traffic police's gesture；
Compare each second probability and the second predetermined threshold value, traffic police's gesture is determined according to comparison result.
5. gesture identification method according to claim 1, which is characterized in that its week in the acquisition vehicle traveling process
Before the obstacle object point cloud enclosed, the method further includes：
Distance of the laser sensor to each barrier peak and the horizontal distance to each barrier are determined respectively；
Each obstacle is determined according to distance of the laser sensor to each barrier peak and the horizontal distance to each barrier
The height of object；
Correspondingly, surrounding obstacle object point cloud includes in the acquisition vehicle traveling process：
Obtain the obstacle object point cloud that surrounding height in vehicle traveling process meets preset condition.
6. gesture identification method according to claim 1, which is characterized in that the method further includes：
According to the gesture identified, corresponding operation is executed.
7. gesture identification method according to claim 1, which is characterized in that in the feature according to each barrier profile
Before data judge whether each barrier is people respectively, the method further includes：
The profile of each barrier is normalized.
8. a kind of gesture identifying device, which is characterized in that including：
Data acquisition module, for obtaining surrounding obstacle object point cloud in vehicle traveling process；
Profile extraction module, for determining each barrier profile according to the obstacle object point cloud of the surrounding；
Barrier classification judgment module, for whether judging each barrier respectively according to the characteristic of each barrier profile
For people；
Gesture recognition module, for when barrier is people, its gesture to be identified according to advance trained traffic police's gesture model.
9. a kind of automobile, which is characterized in that the automobile includes：
Laser sensor for obtaining the obstacle object point cloud around in the vehicle traveling process, and obtains each obstacle distance；
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 are real
The now gesture identification method as described in any in claim 1-7.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
The gesture identification method as described in any in claim 1-7 is realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810288399.6A CN108501954A (en) | 2018-04-03 | 2018-04-03 | A kind of gesture identification method, device, automobile and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810288399.6A CN108501954A (en) | 2018-04-03 | 2018-04-03 | A kind of gesture identification method, device, automobile and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108501954A true CN108501954A (en) | 2018-09-07 |
Family
ID=63379918
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810288399.6A Pending CN108501954A (en) | 2018-04-03 | 2018-04-03 | A kind of gesture identification method, device, automobile and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108501954A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109344804A (en) * | 2018-10-30 | 2019-02-15 | 百度在线网络技术（北京）有限公司 | A kind of recognition methods of laser point cloud data, device, equipment and medium |
| CN109508659A (en) * | 2018-10-31 | 2019-03-22 | 绍兴文理学院 | A kind of face identification system and method for crossing |
| CN110046569A (en) * | 2019-04-12 | 2019-07-23 | 北京百度网讯科技有限公司 | A kind of data processing method, device and electronic equipment |
| CN111695420A (en) * | 2020-04-30 | 2020-09-22 | 华为技术有限公司 | Gesture recognition method and related device |
| CN112560548A (en) * | 2019-09-24 | 2021-03-26 | 北京百度网讯科技有限公司 | Method and apparatus for outputting information |
| US20220153275A1 (en) * | 2020-11-16 | 2022-05-19 | Autobrains Technologies Ltd | Incorporating Human Communication into Driving Related Decisions |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955215A (en) * | 2014-04-15 | 2014-07-30 | 桂林电子科技大学 | Automatic obstacle avoidance trolley based on gesture recognition and control device and method of automatic obstacle avoidance trolley |
| CN104134061A (en) * | 2014-08-15 | 2014-11-05 | 上海理工大学 | Number gesture recognition method for support vector machine based on feature fusion |
| CN105320937A (en) * | 2015-09-25 | 2016-02-10 | 北京理工大学 | Kinect based traffic police gesture recognition method |
| CN106203380A (en) * | 2016-07-20 | 2016-12-07 | 中国科学院计算技术研究所 | Ultrasound wave gesture identification method and system |
| CN106295586A (en) * | 2016-08-16 | 2017-01-04 | 长春理工大学 | Humanoid target identification method based on single line cloud data machine learning and device |
| CN106570454A (en) * | 2016-10-10 | 2017-04-19 | 同济大学 | Pedestrian traffic parameter extraction method based on mobile laser scanning |
| CN106709475A (en) * | 2017-01-22 | 2017-05-24 | 百度在线网络技术（北京）有限公司 | Obstacle recognition method and device, computer equipment and readable storage medium |
| CN106845412A (en) * | 2017-01-20 | 2017-06-13 | 百度在线网络技术（北京）有限公司 | Obstacle recognition method and device, computer equipment and computer-readable recording medium |
| CN106951847A (en) * | 2017-03-13 | 2017-07-14 | 百度在线网络技术（北京）有限公司 | Obstacle detection method, device, equipment and storage medium |
| CN107167788A (en) * | 2017-03-21 | 2017-09-15 | 深圳市速腾聚创科技有限公司 | Obtain laser radar calibration parameter, the method and system of laser radar calibration |
| CN107797666A (en) * | 2017-11-21 | 2018-03-13 | 出门问问信息科技有限公司 | Gesture identification method, device and electronic equipment |
| CN107813817A (en) * | 2016-08-25 | 2018-03-20 | 大连楼兰科技股份有限公司 | Unmanned Systems, unmanned method and vehicle |
-
2018
- 2018-04-03 CN CN201810288399.6A patent/CN108501954A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955215A (en) * | 2014-04-15 | 2014-07-30 | 桂林电子科技大学 | Automatic obstacle avoidance trolley based on gesture recognition and control device and method of automatic obstacle avoidance trolley |
| CN104134061A (en) * | 2014-08-15 | 2014-11-05 | 上海理工大学 | Number gesture recognition method for support vector machine based on feature fusion |
| CN105320937A (en) * | 2015-09-25 | 2016-02-10 | 北京理工大学 | Kinect based traffic police gesture recognition method |
| CN106203380A (en) * | 2016-07-20 | 2016-12-07 | 中国科学院计算技术研究所 | Ultrasound wave gesture identification method and system |
| CN106295586A (en) * | 2016-08-16 | 2017-01-04 | 长春理工大学 | Humanoid target identification method based on single line cloud data machine learning and device |
| CN107813817A (en) * | 2016-08-25 | 2018-03-20 | 大连楼兰科技股份有限公司 | Unmanned Systems, unmanned method and vehicle |
| CN106570454A (en) * | 2016-10-10 | 2017-04-19 | 同济大学 | Pedestrian traffic parameter extraction method based on mobile laser scanning |
| CN106845412A (en) * | 2017-01-20 | 2017-06-13 | 百度在线网络技术（北京）有限公司 | Obstacle recognition method and device, computer equipment and computer-readable recording medium |
| CN106709475A (en) * | 2017-01-22 | 2017-05-24 | 百度在线网络技术（北京）有限公司 | Obstacle recognition method and device, computer equipment and readable storage medium |
| CN106951847A (en) * | 2017-03-13 | 2017-07-14 | 百度在线网络技术（北京）有限公司 | Obstacle detection method, device, equipment and storage medium |
| CN107167788A (en) * | 2017-03-21 | 2017-09-15 | 深圳市速腾聚创科技有限公司 | Obtain laser radar calibration parameter, the method and system of laser radar calibration |
| CN107797666A (en) * | 2017-11-21 | 2018-03-13 | 出门问问信息科技有限公司 | Gesture identification method, device and electronic equipment |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109344804A (en) * | 2018-10-30 | 2019-02-15 | 百度在线网络技术（北京）有限公司 | A kind of recognition methods of laser point cloud data, device, equipment and medium |
| CN109508659A (en) * | 2018-10-31 | 2019-03-22 | 绍兴文理学院 | A kind of face identification system and method for crossing |
| CN110046569A (en) * | 2019-04-12 | 2019-07-23 | 北京百度网讯科技有限公司 | A kind of data processing method, device and electronic equipment |
| CN112560548A (en) * | 2019-09-24 | 2021-03-26 | 北京百度网讯科技有限公司 | Method and apparatus for outputting information |
| CN112560548B (en) * | 2019-09-24 | 2024-04-02 | 北京百度网讯科技有限公司 | Method and device for outputting information |
| CN111695420A (en) * | 2020-04-30 | 2020-09-22 | 华为技术有限公司 | Gesture recognition method and related device |
| CN111695420B (en) * | 2020-04-30 | 2024-03-08 | 华为技术有限公司 | Gesture recognition method and related device |
| US20220153275A1 (en) * | 2020-11-16 | 2022-05-19 | Autobrains Technologies Ltd | Incorporating Human Communication into Driving Related Decisions |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108501954A (en) | A kind of gesture identification method, device, automobile and storage medium | |
| JP6842520B2 (en) | Object detection methods, devices, equipment, storage media and vehicles | |
| CN108960183B (en) | Curve target identification system and method based on multi-sensor fusion | |
| JP7395301B2 (en) | Obstacle detection method, obstacle detection device, electronic equipment, vehicle and storage medium | |
| CN112184818B (en) | Vision-based vehicle positioning method and parking lot management system applying same | |
| CN108537197B (en) | Lane line detection early warning device and method based on deep learning | |
| CN106169244B (en) | Device and method is provided using the guidance information of crossing recognition result | |
| US8379928B2 (en) | Obstacle detection procedure for motor vehicle | |
| CN110163930A (en) | Lane line generation method, device, equipment, system and readable storage medium storing program for executing | |
| CN111860274B (en) | Traffic police command gesture recognition method based on head orientation and upper half skeleton characteristics | |
| CN109101861A (en) | Obstacle identity recognition methods, device, equipment and storage medium | |
| US20100097455A1 (en) | Clear path detection using a vanishing point | |
| CN104778444A (en) | Method for analyzing apparent characteristic of vehicle image in road scene | |
| CN111899515B (en) | Vehicle detection system based on wisdom road edge calculates gateway | |
| CN107389084A (en) | Planning driving path planing method and storage medium | |
| CN104915642B (en) | Front vehicles distance measuring method and device | |
| CN111967396A (en) | Processing method, device and equipment for obstacle detection and storage medium | |
| Thakur et al. | Deep learning-based parking occupancy detection framework using ResNet and VGG-16 | |
| CN111739333B (en) | Empty parking space identification method | |
| US11790683B2 (en) | Gesture based authentication for autonomous vehicles | |
| Ghahremannezhad et al. | Automatic road detection in traffic videos | |
| JP6472504B1 (en) | Information processing apparatus, information processing program, and information processing method | |
| Coronado et al. | Detection and classification of road signs for automatic inventory systems using computer vision | |
| CN113673527A (en) | License plate recognition method and system | |
| Guo et al. | Visibility detection based on the recognition of the preceding vehicle’s taillight signals |
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 |
Application publication date: 20180907 |
|
| RJ01 | Rejection of invention patent application after publication |