CN109800680A - A kind of method and device of object in determining video - Google Patents
A kind of method and device of object in determining video Download PDFInfo
- Publication number
- CN109800680A CN109800680A CN201811646178.8A CN201811646178A CN109800680A CN 109800680 A CN109800680 A CN 109800680A CN 201811646178 A CN201811646178 A CN 201811646178A CN 109800680 A CN109800680 A CN 109800680A
- Authority
- CN
- China
- Prior art keywords
- image
- region
- images
- recognized
- face
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000001514 detection method Methods 0.000 claims abstract description 59
- 238000012545 processing Methods 0.000 claims abstract description 28
- 238000012806 monitoring device Methods 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 12
- 241000406668 Loxodonta cyclotis Species 0.000 claims 1
- 230000001815 facial effect Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The embodiment of the present application provides a kind of method and device of the object in determining video, it is related to technical field of video processing, when obtaining the first images to be recognized, first judge whether the first images to be recognized is detection frame image, if it is, object detection then is carried out to the first images to be recognized, determines corresponding first image-region of each object in the first images to be recognized.Then according to the corresponding third image-region of each second object in the second images to be recognized, predict each second object corresponding second image-region in the first images to be recognized, when the second image-region of the second object and the first image-region of the first object meet and impose a condition, it determines the second object and the first object is same target, the image-region of same target is determined according to the first image-region and the second image-region, on the one hand avoid same target there are two image-regions in an images to be recognized, it ensure that continuity of the same target in each images to be recognized simultaneously.
Description
Technical field
The present embodiments relate to a kind of method of the object in technical field of video processing more particularly to determining video and
Device.
Background technique
In today's society, for the needs of safeguard management, it is various public that monitoring device has been covered with street, community, building etc.
Occasion.When alert occurs, the image of suspect or suspected vehicles is first determined from the video data that monitoring device acquires, so
Suspect or suspected vehicles are tracked down and arrested according to the image of suspect or suspected vehicles by police afterwards.
In the prior art, monitoring device acquisition video flowing after, mostly use to every frame image in video flowing carry out detection and
Identification, determines the objects such as face or the vehicle in every frame image, then again matches the object detected in each frame image,
Determine all images an of object, this method is imitated due to needing to be detected and identified that calculation amount is larger to every frame image
Rate is low.
Summary of the invention
Due in the prior art, when determining the image of same target from video flowing, every frame image being detected and being known
Not, the problem that calculation amount is larger and efficiency is lower, the embodiment of the present application provide the object in a kind of determining video method and
Device.
On the one hand, the embodiment of the present application provides a kind of method of the object in determining video, comprising:
When determining the first images to be recognized is detection frame image, object detection is carried out to first images to be recognized,
Determine corresponding first image-region of each first object in first images to be recognized；
According to the corresponding third image-region of each second object in the second images to be recognized, each second object is predicted
Corresponding second image-region in first images to be recognized；First images to be recognized and second figure to be identified
Adjacent image as being monitoring device acquisition；
When the second image-region of the second object and the first image-region of the first object meet and impose a condition, institute is determined
It states the second object and first object is same target.
Optionally it is determined that second object and first object is after same targets, further includes:
It is second object that first object is marked in first images to be recognized, so as to described second pair
As being tracked.
Optionally, the first image-region of the second image-region of second object and the first object meets setting item
Part, comprising:
There are intersections for first image-region of the second image-region of the second object and the first object；Or
The locational space of first image-region of the locational space and the first object of the second image-region of the second object it
Between distance be less than given threshold.
Optionally, the determination second object and first object is after same targets, further includes:
The image-region of the same target is determined according to the first image region and second image-region.
On the one hand, the embodiment of the present application provides a kind of device of the object in determining video, comprising:
Detection module, for determine the first images to be recognized be detection frame image when, to first images to be recognized
Object detection is carried out, determines corresponding first image-region of each first object in first images to be recognized；
Prediction module, for according to the corresponding third image-region of each second object in the second images to be recognized, prediction
Each second object corresponding second image-region in first images to be recognized；First images to be recognized and institute
State the adjacent image that the second images to be recognized is monitoring device acquisition；
Processing module is set for meeting in the second image-region of the second object and the first image-region of the first object
When condition, determines second object and first object is same target.
Optionally, the processing module is also used to:
Second object and first object are determined to be got the bid after same target in first images to be recognized
Remember that first object is second object, to track to second object.
Optionally, the processing module is specifically used for:
In the second image-region and the first image-region of the first object of the second object, there are when intersection, determine described
Two objects and first object are same target；Or
In the locational space of the first image-region of the locational space and the first object of the second image-region of the second object
The distance between when being less than given threshold, determine that second object and first object are same target.
Optionally, the processing module is also used to:
After determining that second object and first object are same target, according to the first image region and institute
State the image-region that the second image-region determines the same target.
On the one hand, the embodiment of the present application provides a kind of equipment of the object in determining video, including at least one processing
Unit and at least one storage unit, wherein the storage unit is stored with computer program, when described program is described
When processing unit executes, so that the step of processing unit executes the method for determining the object in video.
On the one hand, the embodiment of the present application provides a kind of computer-readable medium, and being stored with can be by determining in video
The computer program that the equipment of object executes, when described program is run in the equipment for determining the object in video, so that institute
The equipment for stating the object in determining video executes the step of method for determining the object in video.
In the embodiment of the present application, since the image in the video of monitoring device acquisition is first divided into detection frame image and prediction
Frame image, therefore when obtaining the first images to be recognized, first judge whether the first images to be recognized is detection frame image, if so,
Object detection then is carried out to the first images to be recognized, determines corresponding first image-region of each object in the first images to be recognized.
Then according to the corresponding third image-region of each second object in the second images to be recognized, predict each second object first to
Corresponding second image-region in image is identified, in the second image-region of the second object and the first image-region of the first object
When meeting setting condition, determine that the second object and the first object are same target, according to the first image-region and the second image district
Domain determines the image-region of same target, on the one hand avoids same target and occurs two image districts in an images to be recognized
Domain, while ensure that continuity of the same target in each images to be recognized.Determining the first images to be recognized for prediction frame
When image, according to the corresponding third image-region of each second object in the second images to be recognized, predict each second object the
Corresponding second image-region in one images to be recognized, without every frame images to be recognized is detected and is identified, thus
Calculation amount is reduced, while improving efficiency.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill in field, without any creative labor, it can also be obtained according to these attached drawings
His attached drawing.
Fig. 1 is the system architecture diagram that the embodiment of the present application is applicable in；
Fig. 2 is the flow diagram of the method for the object in a kind of determining video of the embodiment of the present application；
Fig. 3 is the structural schematic diagram of the device of the object in a kind of determining video of the embodiment of the present application；
Fig. 4 is the structural schematic diagram of the equipment of the object in a kind of determining video of the embodiment of the present application.
Specific embodiment
In order to which the purpose of the present invention, technical solution and beneficial effect is more clearly understood, below in conjunction with attached drawing and implementation
Example, the present invention will be described in further detail.It should be appreciated that specific embodiment described herein is only used to explain this hair
It is bright, it is not intended to limit the present invention.
Fig. 1 illustrates the applicable system architecture of the embodiment of the present application, includes monitoring device in the system architecture
101, server 102.Monitoring device 101 acquires video in real time, and the video of acquisition is then sent to server 102, server
It include the device for determining the object in video in 102, server 102 obtains image from video, then determines pair in image
As corresponding image-region, image-region of the same target in video in each frame image is determined again later, it is subsequent using same
Image-region of the object in each frame image is filed or is identified to object.Monitoring device 101 passes through wireless network and clothes
Business device 102 connects, and monitoring device is the electronic equipment, such as camera, video camera, video recorder etc. for having acquisition image function.
Server 102 is the server cluster or cloud computing center of a server or several servers composition.
Based on system architecture shown in FIG. 1, the embodiment of the present application provides a kind of method of the object in determining video
Process, the process of this method can be executed by determining the device of the object in video, determine that the device of the object in video can be with
It is server 102 shown in FIG. 1, as shown in Figure 2, comprising the following steps:
Step S201 carries out object to the first images to be recognized when determining the first images to be recognized is detection frame image
Detection determines corresponding first image-region of each first object in the first images to be recognized.
First images to be recognized is acquired by monitoring device, and monitoring device acquires video flowing in real time, and video flowing is by more
Frame images to be recognized composition, is in advance marked the images to be recognized in video flowing.
It is detection frame figure by the image tagged for needing to carry out object detection in video flowing in a kind of possible embodiment
Picture.Illustratively, setting includes 10 frame images in one section of video flowing, is detection frame by first frame image and the 5th frame image tagged
Image.
It is detection frame figure by the image tagged for needing to carry out object detection in video flowing in a kind of possible embodiment
The image tagged for needing to carry out object prediction is prediction frame image by picture.Illustratively, setting includes 10 frames in one section of video flowing
First frame image and the 5th frame image tagged are detection frame image by image, and the second frame image to the 4th frame image tagged is pre-
Frame image is surveyed, the 6th frame to the tenth frame flag is prediction frame image.
Object can be the objects such as face or vehicle, and image-region can be the frame with regular shape, or not have
The region of regular shape.
Illustratively, Face datection is carried out to the first images to be recognized, determines that each face is corresponding in the first images to be recognized
Face frame.
Illustratively, vehicle detection is carried out to the first images to be recognized, determines each vehicle pair in the first images to be recognized
The vehicle frame answered.
Optionally, when determining the first images to be recognized for prediction frame image, according to each the in the second images to be recognized
The corresponding third image-region of two objects predicts each second object corresponding second image district in first images to be recognized
Domain, the first images to be recognized and the second images to be recognized are the adjacent image of monitoring device acquisition.
Step S202 predicts each the according to the corresponding third image-region of each second object in the second images to be recognized
Two objects corresponding second image-region in the first images to be recognized.
First images to be recognized and the second images to be recognized are the adjacent image of monitoring device acquisition, specifically,
Second images to be recognized can be detection frame image or non-detection frame image.
Step S203 meets in the second image-region of the second object and the first image-region of the first object and imposes a condition
When, it determines the second object and the first object is same target.
In a kind of possible embodiment, in the second image-region of the second object and the first image district of the first object
There are when intersection, determine the second object and the first object for same target in domain.
Illustratively, setting carries out Face datection to the first images to be recognized, determines the first face frame of face A, according to
The corresponding third face frame of each second face in second images to be recognized predicts each second face in the first images to be recognized
Corresponding second face frame.Set in each second face frame that there are the presence of the first face frame of a second face frame and face A
Face in the second face frame is then determined as face A by intersection.
In a kind of possible embodiment, in the locational space and the first object of the second image-region of the second object
When the distance between locational space of first image-region is less than given threshold, determines the second object and the first object is with a pair of
As.
Since the image in the video of monitoring device acquisition is first divided into detection frame image and prediction frame image, therefore obtaining
When the first images to be recognized, first judge whether the first images to be recognized is detection frame image, if it is, to the first figure to be identified
As carrying out object detection, corresponding first image-region of each object in the first images to be recognized is determined.Then according to second wait know
The corresponding third image-region of each second object in other image predicts that each second object is corresponding in the first images to be recognized
Second image-region, when the second image-region of the second object and the first image-region of the first object meet and impose a condition,
It determines that the second object and the first object are same target, on the one hand avoids same target and occur two in an images to be recognized
A image-region, while ensure that continuity of the same target in each images to be recognized.
Optionally, optionally, corresponding for any second object in the second images to be recognized in above-mentioned steps S202
Third image-region, determine that the similarity of third image-region corresponding with the second object is big from the first images to be recognized
In the image-region of preset threshold, as the second object in the first images to be recognized corresponding second image-region.
Illustratively, the second object is set as face, it is known that the corresponding face frame of face in the second images to be recognized, for
The corresponding face frame of each face, is determined similar to the face frame from the first images to be recognized in second images to be recognized
Degree is greater than the face frame of preset threshold, as face frame of the face in the first images to be recognized.
In a kind of possible embodiment, the apparent area of feature is selected from the corresponding third image-region of the second object
The region is compared with the first images to be recognized, the similarity with the region is determined from the first images to be recognized by domain
It is corresponding in the first images to be recognized as the second object after then expanding image-region greater than the image-region of preset threshold
Image-region.
Illustratively, the second object is set as face, it is known that the corresponding face frame of the second face in the second images to be recognized,
For the face frame in the second images to be recognized, the apparent region of feature in face frame, such as mouth and beard are chosen first,
The apparent region of feature in face frame is compared with the first images to be recognized, determines and is somebody's turn to do from the first images to be recognized
The similarity in region is greater than the image-region of preset threshold, to be identified first as the face after then expanding image-region
Corresponding face frame in image.
Further, in a kind of possible embodiment, if the image-region that similarity is greater than preset threshold is more than one
It is a, then choose the highest image-region of similarity as the second object corresponding second image district in the first images to be recognized
Domain corresponds to multiple images region in an images to be recognized so as to avoid an object.
In a kind of possible embodiment, if the image-region that similarity is greater than preset threshold is more than one, choose
The nearest image-region in the position of the corresponding image-region of the second object, exists as the second object in the second images to be recognized of distance
Corresponding second image-region in first images to be recognized.
Since the time interval between the images to be recognized of monitoring device acquisition is shorter, therefore in adjacent two images to be recognized
The mobile distance of same target is shorter, and the third image-region of the second object in the second images to be recognized of prediction is first wait know
When the second image-region in other image, image-region that can be nearest with selected distance third image-region, as the second object
Corresponding second image-region in the first images to be recognized, it is corresponding in an images to be recognized so as to avoid an object
Multiple images region.
Optionally, each second object is predicted in the first images to be recognized after corresponding second image-region, for appointing
One second object detects the key point of the second object in the second image-region of the second object.According to key point adjustment second
Object corresponding second image-region in the first images to be recognized.
Specifically, the key point of the second object is the key point of the second object for identification, for example the key point of face is
Eyes, nose, mouth, ear, hair etc..The key point of vehicle includes license plate, vehicle window, wheel etc..
Illustratively, set the second object as face, predict face in the first images to be recognized after corresponding face frame,
The key point in face frame is detected, if including two eyes, a nose in the key point of monitoring, is illustrated in the face frame
Face is face front, and mouth is not included in face frame, then can expand face frame downwards, make include in face frame
The mouth of face.
Due to detecting the pass of the second object after predicting the second image-region of second object in the first images to be recognized
Key point is then based on the image-region that key point adjusts the second object, to keep the image-region of the second object more acurrate.
Optionally, after determining that the second object and the first object are same target, can according to the first image-region and
Second image-region determines the image-region of same target.
In a kind of possible embodiment, will the first image-region and the second image-region merge after obtained image district
Image-region of the domain as the same target.
Since the first image-region and the second image-region are merged into an image-region, so that same target be avoided to exist
There is multiple images region in one images to be recognized, while ensure that the continuity of the image-region of same target.
In a kind of possible embodiment, the first image-region and the second image-region are merged and determine pre- merging image
Region, then the key point of the pre- object for merging image-region of detection, then adjusts according to key point and merges image-region in advance, really
Determine the image-region of same target.
It optionally, can be in the first images to be recognized after determining the second object and the first object for same target
The first object of label is the second object, to track to the second object.
Optionally, after the image-region that same target is determined according to the first image-region and the second image-region, according to
Image-region of the same target in different images to be recognized determines the identification image of object.
Illustratively, setting same face includes 10 image-regions in 10 frame images to be recognized, then schemes this 10
10 identification images as region as face.When the subsequent identification image to the face is filed, this 10 can be known
Other image co-registration is 1 archival images, which is compared with face archives in facial image database, is determined matched
This 10 identification images are included into matched face archives again later by face archives.
Optionally, when object is face, the figure of same target is determined according to the first image-region and the second image-region
As that can determine whether the second image-region is facial image, specifically includes following several embodiments after region:
In a kind of possible embodiment, the face key point in the second image-region is detected, according to face key point
Determine whether the second image-region is facial image.
Face key point includes eyes, nose, mouth, ear, hair etc..
The face key point in the second image-region is detected, whether judges the face key point for including in the second image-region
Meet preset condition, if so, the second image-region is determined as facial image, is otherwise determined as the second image-region inhuman
Face image.
In a kind of possible embodiment, use face classification device determine the second image-region whether for facial image,
Face classification device is determined after being trained to artificial neural network using preselecting the other image of marking class as training sample.
The result of face classification device output includes facial image and inhuman face image.
In a kind of possible embodiment, the face key point and picture quality in the second image-region are detected.By
Two image-regions input face classification device, determine that the classification results of the second image-region, face classification device are to preselect marking class
Other image is training sample, is determined after being trained to artificial neural network.It is closed according to the face in the second image-region
The classification results of key point, the picture quality of the second image-region and the second image-region determine whether the second image-region is people
Face image.
Specifically, the classification results of face classification device output include facial image and inhuman face image.It in advance can be root
The classification results and picture quality of the classification results, face classification device that determine according to face key point distribute weight, then basis
Weight merges the classification results of classification results, face classification device that face key point determines and picture quality, determines
Whether the second image-region is facial image.
Due to first judging whether the second image-region is facial image, then having determined that as people according to same face object
Second image-region of face image, determines the identification image of face object, so that the precision of the identification image of face object is improved,
Further increase the accuracy that the subsequent identification image using face object is filed.
Optionally, for any object, the key point of test object in the image-region of object.It is adjusted according to key point
Object corresponding image-region in the first images to be recognized.
Specifically, the key point of object is the key point of object for identification, for example the key point of face is eyes, nose
Son, mouth, ear, hair etc..The key point of vehicle includes license plate, vehicle window, wheel etc..
Illustratively, the key point in face frame is detected, if in the key point of detection including two eyes, a nose,
Then illustrate that the face in the face frame is face front, and mouth is not included in face frame, then can expand people downwards
Face frame makes the mouth in face frame comprising face.
Due to the key point of test object, it is then based on key point adjustment image-region, to keep image-region more acurrate.
Embodiment in order to preferably explain the present invention describes the embodiment of the present application below with reference to specific implement scene and provides
A kind of determining video in object method, this method can by determine video in object device execute.
Setting includes 10 frame images to be recognized in video flowing, and wherein first frame images to be recognized is detection frame image, third
Frame images to be recognized is detection frame image, remaining is prediction frame image.Face inspection is carried out to first frame images to be recognized first
It surveys, determines the first face detection block of each face in first frame images to be recognized.For face A in first frame images to be recognized
The first face detection block, first the key point of face A in the first face detection block is detected, according to the key point tune of detection
Whole first face detection block.Whether the first face detection block for then judging face A is facial image, if so, prediction face A
The first face detection block in the corresponding second face prediction block of the second frame images to be recognized.To face in the second face prediction block
The key point of A is detected, and adjusts the second face prediction block according to the key point of detection.Judge the second people of face A again later
Whether face prediction block is facial image, if so, the second face prediction block of prediction face A is right in third frame images to be recognized
The third face prediction block answered detects the key point of face A in third face prediction block, according to the key point tune of detection
Whole third face prediction block.Face datection is carried out to third frame images to be recognized simultaneously, determines the of third frame images to be recognized
Three Face datection frames.There are a third face detection block and people in the third face detection block of setting third frame images to be recognized
There are intersections for the third face prediction block of face A, then using the third face prediction block of third face detection block amendment face A.
Whether the third face prediction block for judging revised face A again later is facial image, if so, the face after forecast value revision
The third face prediction block of A is in the corresponding 4th face prediction block of the 4th frame images to be recognized.And so on, until that cannot predict
Face A face prediction block in next frame images to be recognized.It sets in 10 frame images to be recognized, the corresponding 8 face frames of face A,
Then using this 8 face frames as the identification image of face A, face A identify or file face A for subsequent.
In addition in the third face detection block of third frame images to be recognized, as the third party with third frame images to be recognized
When face prediction block does not have intersection, such third face detection block is determined as the corresponding Face datection frame of a new face.It is right
Key point in third face detection block is detected, and adjusts third face detection block according to the key point of detection.Then judge
Whether third face detection block is facial image, if so, prediction third face detection block is corresponding in the 4th frame images to be recognized
The 4th face prediction block, and so on, until that cannot predict the face prediction block in next frame images to be recognized.
In the embodiment of the present application, since the image in the video of monitoring device acquisition is first divided into detection frame image and prediction
Frame image, therefore when obtaining the first images to be recognized, first judge whether the first images to be recognized is detection frame image, if so,
Object detection then is carried out to the first images to be recognized, determines corresponding first image-region of each object in the first images to be recognized.
Then according to the corresponding third image-region of each second object in the second images to be recognized, predict each second object first to
Corresponding second image-region in image is identified, in the second image-region of the second object and the first image-region of the first object
When meeting setting condition, determine that the second object and the first object are same target, according to the first image-region and the second image district
Domain determines the image-region of same target, on the one hand avoids same target and occurs two image districts in an images to be recognized
Domain, while ensure that continuity of the same target in each images to be recognized.Determining the first images to be recognized for prediction frame
When image, according to the corresponding third image-region of each second object in the second images to be recognized, predict each second object the
Corresponding second image-region in one images to be recognized, without every frame images to be recognized is detected and is identified, thus
Calculation amount is reduced, while improving efficiency.
Based on the same technical idea, the embodiment of the invention provides a kind of device of the object in determining video, such as Fig. 3
Shown, which includes:
Detection module, for determine the first images to be recognized be detection frame image when, to first images to be recognized
Object detection is carried out, determines corresponding first image-region of each first object in first images to be recognized；
Prediction module, for according to the corresponding third image-region of each second object in the second images to be recognized, prediction
Each second object corresponding second image-region in first images to be recognized；First images to be recognized and institute
State the adjacent image that the second images to be recognized is monitoring device acquisition；
Processing module is set for meeting in the second image-region of the second object and the first image-region of the first object
When condition, determines second object and first object is same target.
Optionally, the processing module is also used to:
Second object and first object are determined to be got the bid after same target in first images to be recognized
Remember that first object is second object, to track to second object.
Optionally, the processing module is specifically used for:
In the second image-region and the first image-region of the first object of the second object, there are when intersection, determine described
Two objects and first object are same target；Or
In the locational space of the first image-region of the locational space and the first object of the second image-region of the second object
The distance between when being less than given threshold, determine that second object and first object are same target.
Optionally, the processing module is also used to:
After determining that second object and first object are same target, according to the first image region and institute
State the image-region that the second image-region determines the same target.
Based on the same technical idea, the embodiment of the present application provides a kind of equipment of the object in determining video, such as Fig. 4
It is shown, including at least one processor 401, and the memory 402 being connect at least one processor, in the embodiment of the present application
The specific connection medium between processor 401 and memory 402 is not limited, is led between processor 401 and memory 402 in Fig. 4
It crosses for bus connection.Bus can be divided into address bus, data/address bus, control bus etc..
In the embodiment of the present application, memory 402 is stored with the instruction that can be executed by least one processor 401, at least
One processor 401, which passes through, executes the instruction that memory 402 stores, in the method that can execute the object in aforementioned determining video
Included step.
Wherein, processor 401 is the control centre of the equipment of the object in determining video, can use various interfaces and line
Road connects the various pieces for determining the equipment of the object in video, by running or executing the instruction being stored in memory 402
And the data being stored in memory 402 are called, so that it is determined that the object in video.Optionally, processor 401 may include one
A or multiple processing units, processor 401 can integrate application processor and modem processor, wherein application processor master
Processing operation system, user interface and application program etc. are wanted, modem processor mainly handles wireless communication.It is understood that
It is that above-mentioned modem processor can not also be integrated into processor 401.In some embodiments, processor 401 and storage
Device 402 can realize that in some embodiments, they can also be realized respectively on independent chip on the same chip.
Processor 401 can be general processor, such as central processing unit (CPU), digital signal processor, dedicated integrated
Circuit (Application Specific Integrated Circuit, ASIC), field programmable gate array or other can
Perhaps transistor logic, discrete hardware components may be implemented or execute the application implementation for programmed logic device, discrete gate
Each method, step and logic diagram disclosed in example.General processor can be microprocessor or any conventional processor
Deng.The step of method in conjunction with disclosed in the embodiment of the present application, can be embodied directly in hardware processor and execute completion, Huo Zheyong
Hardware and software module combination in processor execute completion.
Memory 402 is used as a kind of non-volatile computer readable storage medium storing program for executing, can be used for storing non-volatile software journey
Sequence, non-volatile computer executable program and module.Memory 402 may include the storage medium of at least one type,
It such as may include flash memory, hard disk, multimedia card, card-type memory, random access storage device (Random Access
Memory, RAM), static random-access memory (Static Random Access Memory, SRAM), may be programmed read-only deposit
Reservoir (Programmable Read Only Memory, PROM), read-only memory (Read Only Memory, ROM), band
Electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory,
EEPROM), magnetic storage, disk, CD etc..Memory 402 can be used for carrying or storing have instruction or data
The desired program code of structure type and can by any other medium of computer access, but not limited to this.The application is real
Applying the memory 402 in example can also be circuit or other devices that arbitrarily can be realized store function, for storing program
Instruction and/or data.
Based on the same inventive concept, the embodiment of the present application provides a kind of computer-readable medium, and being stored with can be by true
Determine the computer program that the equipment of the object in video executes, when program is run in the equipment for determining the object in video,
So that the equipment for determining the object in video executes the step of method for determining the object in video.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method or computer program product.
Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the present invention
Form.It is deposited moreover, the present invention can be used to can be used in the computer that one or more wherein includes computer usable program code
The shape for the computer program product implemented on storage media (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (10)
1. a kind of method of the object in determining video characterized by comprising
When determining the first images to be recognized is detection frame image, object detection is carried out to first images to be recognized, is determined
Corresponding first image-region of each first object in first images to be recognized；
According to the corresponding third image-region of each second object in the second images to be recognized, predict each second object in institute
State corresponding second image-region in the first images to be recognized；First images to be recognized and second images to be recognized are
The adjacent image of monitoring device acquisition；
When the second image-region of the second object and the first image-region of the first object meet and impose a condition, described the is determined
Two objects and first object are same target.
2. the method as described in claim 1, which is characterized in that determination second object and first object are same
After an object, further includes:
Marked in first images to be recognized first object be second object, so as to second object into
Line trace.
3. the method as described in claim 1, which is characterized in that the second image-region of second object and the first object
First image-region, which meets, to impose a condition, comprising:
There are intersections for first image-region of the second image-region of the second object and the first object；Or
Between the locational space of first image-region of the locational space and the first object of the second image-region of the second object
Distance is less than given threshold.
4. the method as described in claim 1, which is characterized in that determination second object and first object are same
After an object, further includes:
The image-region of the same target is determined according to the first image region and second image-region.
5. a kind of device of the object in determining video characterized by comprising
Detection module, for being carried out to first images to be recognized when determining the first images to be recognized is detection frame image
Object detection determines corresponding first image-region of each first object in first images to be recognized；
Prediction module, for according to the corresponding third image-region of each second object in the second images to be recognized, described in prediction
Each second object corresponding second image-region in first images to be recognized；First images to be recognized and described
Two images to be recognized are the adjacent image of monitoring device acquisition；
Processing module imposes a condition for meeting in the second image-region of the second object and the first image-region of the first object
When, it determines second object and first object is same target.
6. device as claimed in claim 5, which is characterized in that the processing module is also used to:
Second object and first object are determined to mark institute in first images to be recognized after same target
Stating the first object is second object, to track to second object.
7. device as claimed in claim 5, which is characterized in that the processing module is specifically used for:
In the second image-region and the first image-region of the first object of the second object, there are when intersection, determine described second pair
As being same target with first object；Or
Between the locational space of the first image-region of the locational space and the first object of the second image-region of the second object
Distance when being less than given threshold, determine that second object and first object are same target.
8. device as claimed in claim 5, which is characterized in that the processing module is also used to:
After determining that second object and first object are same target, according to the first image region and described the
Two image-regions determine the image-region of the same target.
9. a kind of equipment of the object in determining video, which is characterized in that including at least one processing unit and at least one
Storage unit, wherein the storage unit is stored with computer program, when described program is executed by the processing unit, makes
Obtain the step of processing unit perform claim requires 1~4 any claim the method.
10. a kind of computer-readable medium, which is characterized in that it is stored with can be by the equipment execution of the object in determining video
Computer program, when described program is run in the equipment for determining the object in video, so that pair in the determining video
The equipment perform claim of elephant requires the step of 1~4 any the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811646178.8A CN109800680A (en) | 2018-12-29 | 2018-12-29 | A kind of method and device of object in determining video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811646178.8A CN109800680A (en) | 2018-12-29 | 2018-12-29 | A kind of method and device of object in determining video |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109800680A true CN109800680A (en) | 2019-05-24 |
Family
ID=66556865
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811646178.8A Pending CN109800680A (en) | 2018-12-29 | 2018-12-29 | A kind of method and device of object in determining video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109800680A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107145816A (en) * | 2017-02-24 | 2017-09-08 | 北京悉见科技有限公司 | Object identifying tracking and device |
| CN108229308A (en) * | 2017-11-23 | 2018-06-29 | 北京市商汤科技开发有限公司 | Recongnition of objects method, apparatus, storage medium and electronic equipment |
| CN108986138A (en) * | 2018-05-24 | 2018-12-11 | 北京飞搜科技有限公司 | Method for tracking target and equipment |
-
2018
- 2018-12-29 CN CN201811646178.8A patent/CN109800680A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107145816A (en) * | 2017-02-24 | 2017-09-08 | 北京悉见科技有限公司 | Object identifying tracking and device |
| CN108229308A (en) * | 2017-11-23 | 2018-06-29 | 北京市商汤科技开发有限公司 | Recongnition of objects method, apparatus, storage medium and electronic equipment |
| CN108986138A (en) * | 2018-05-24 | 2018-12-11 | 北京飞搜科技有限公司 | Method for tracking target and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110869936B (en) | Method and system for distributed learning and adaptation in an autonomous vehicle | |
| CN109740517A (en) | A kind of method and device of determining object to be identified | |
| CN109087510B (en) | Traffic monitoring method and device | |
| CN109800679A (en) | A kind of method and device of the attribute information of determining object to be identified | |
| CN110753892A (en) | Method and system for instant object tagging via cross-modality verification in autonomous vehicles | |
| CN109887283B (en) | Road congestion prediction method, system and device based on checkpoint data | |
| CN109815842A (en) | A kind of method and device of the attribute information of determining object to be identified | |
| CN110753953A (en) | Method and system for object-centric stereo vision in autonomous vehicles via cross-modality verification | |
| CN108038176B (en) | Method and device for establishing passerby library, electronic equipment and medium | |
| CN111507278B (en) | Method and device for detecting roadblock and computer equipment | |
| CN109800675A (en) | A kind of method and device of the identification image of determining face object | |
| WO2022156553A1 (en) | Method and apparatus for presenting road information | |
| Kejriwal et al. | Vehicle detection and counting using deep learning basedYOLO and deep SORT algorithm for urban traffic management system | |
| CN117197796A (en) | Vehicle shielding recognition method and related device | |
| CN109800678A (en) | The attribute determining method and device of object in a kind of video | |
| CN109784238A (en) | A kind of method and device of determining object to be identified | |
| CN112418303B (en) | Training method and device for identifying state model and computer equipment | |
| CN112562315B (en) | Method, terminal and storage medium for acquiring traffic flow information | |
| CN109800680A (en) | A kind of method and device of object in determining video | |
| WO2023144780A1 (en) | Crowdsourcing techniques to deploy artificial intelligence systems | |
| CN109800685A (en) | The determination method and device of object in a kind of video | |
| CN109815840A (en) | A kind of method and device of determining identification information | |
| CN110659384B (en) | Video structured analysis method and device | |
| Sooraj et al. | Real-time traffic counter using mobile devices | |
| CN113470009A (en) | Illegal umbrella opening detection and identification method and device, electronic equipment and storage medium |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200506 Address after: Room 01 and 02, 1st floor, No. 180 Yizhou Road, Xuhui District, Shanghai Applicant after: SHANGHAI YITU TECHNOLOGY Co.,Ltd. Address before: Room 01 and 02, 1st floor, No. 180 Yizhou Road, Xuhui District, Shanghai Applicant before: SHANGHAI YITU TECHNOLOGY Co.,Ltd. Applicant before: SHENZHEN YITU INFORMATION TECHNOLOGY Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190524 |
|
| RJ01 | Rejection of invention patent application after publication |