CN110334650A - Object detecting method, device, electronic equipment and storage medium - Google Patents
Object detecting method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110334650A CN110334650A CN201910599369.1A CN201910599369A CN110334650A CN 110334650 A CN110334650 A CN 110334650A CN 201910599369 A CN201910599369 A CN 201910599369A CN 110334650 A CN110334650 A CN 110334650A
- Authority
- CN
- China
- Prior art keywords
- video frame
- target object
- region
- object region
- examined
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/48—Matching video sequences
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The present disclosure discloses a kind of object detecting method, device, electronic equipment and storage mediums.This method comprises: determining the target object region in initial video frame comprising examined object;The traversal video from the initial video frame determines the target object region in current video frame according to the target object region in previous video frame, until the probability that the target object region in current video frame includes examined object is greater than setting targets threshold;According to the target object region of current video frame, object detection result is determined.The speed of object detection can be improved in the embodiment of the present disclosure, and realizes and take into account Detection accuracy and detection speed.
Description
Technical field
The embodiment of the present disclosure be related to data processing technique more particularly to a kind of object detecting method, device, electronic equipment and
Storage medium.
Background technique
With the development of image recognition technology, the recognition accuracy of image recognition technology is higher and higher.For example, can choose
One object to be detected, for example, animal, personage or furniture etc., carry out object to an image using image-recognizing method
The detection of body obtains the location information of the target object in the image.
Currently, the mode for detecting target object in video is usual are as follows: multiple video frames are divided video into, by each view
Frequency frame is respectively adopted image-recognizing method and is detected as single image, to obtain the position of target object in each image
Information.
If target object is static object, when using above-mentioned target object detection method, to video frame each in video
Whole graph region within the scope of detected, in the detection process physical presence largely repeat detection work, cause in video into
The data operation quantity of row detection being related to is big, and occupied space is big, and required operation resource is big.
Summary of the invention
The embodiment of the present disclosure provides a kind of object detecting method, device, electronic equipment and storage medium, and object can be improved
The speed of detection, and realize and take into account Detection accuracy and detection speed.
In a first aspect, the embodiment of the present disclosure provides a kind of object detecting method, this method comprises:
Determine the target object region in initial video frame comprising examined object;
The traversal video from the initial video frame, according to the target object region in previous video frame, forward sight is worked as in determination
Target object region in frequency frame is set until the probability that the target object region in current video frame includes examined object is greater than
Set the goal threshold value;
According to the target object region of current video frame, object detection result is determined.
Second aspect, the embodiment of the present disclosure additionally provide a kind of article detection device, which includes:
Initial video frame object detection module, for determining the target object area in initial video frame comprising examined object
Domain;
Subsequent video frame object detection module, for the traversal video from the initial video frame, according to previous video frame
In target object region, determine the target object region in current video frame, until current video frame in target object area
Domain includes that the probability of examined object is greater than setting targets threshold;
Object detection result determining module determines object detection knot for the target object region according to current video frame
Fruit.
The third aspect, the embodiment of the present disclosure additionally provide a kind of electronic equipment, which includes:
One or more processors;
Memory, 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 object detecting method as described in the embodiment of the present disclosure.
Fourth aspect, the embodiment of the present disclosure additionally provide a kind of computer readable storage medium, are stored thereon with computer
Program realizes the object detecting method as described in the embodiment of the present disclosure when program is executed by processor.
The embodiment of the present disclosure is by beginning stepping through video from initial video frame, according to the target object region of previous video frame
The target object region for determining current video frame, reduces the detection range of each video frame, to reduce the testing number of each video frame
According to amount, while detection data amount reduces the raising that can bring detection speed;In addition, including object to be detected in target object region
When the probability of body is greater than setting targets threshold, determine object detection result, guarantee only determine target object region include to
When detection object, object detection result is just determined, to guarantee the accuracy rate of object detection, hereby it is achieved that taking into account target object
The speed and accuracy rate of detection solve and carry out detecting within the scope of full figure in the prior art to each video frame in video,
Lead to the low efficiency of object detection, the big problem of required operation resource is realized in the successive video frames passed through in video gradually
Precision target object area reduces the data volume of the image procossing of each video frame, improves the speed of target object detection, together
When, it takes into account and detects accurate target object region, to realize the speed and accuracy rate for taking into account object detection, and improve object
The efficiency that physical examination is surveyed.
Detailed description of the invention
Fig. 1 is the flow chart of one of the embodiment of the present disclosure one object detecting method;
Fig. 2 a is the flow chart of one of the embodiment of the present disclosure two object detecting method;
Fig. 2 b is the schematic diagram of one of the embodiment of the present disclosure two current video frame;
Fig. 3 is the structural schematic diagram of one of the embodiment of the present disclosure three article detection device;
Fig. 4 is the structural schematic diagram of one of the embodiment of the present disclosure four electronic equipment.
Specific embodiment
Embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although showing the certain of the disclosure in attached drawing
Embodiment, it should be understood that, the disclosure can be realized by various forms, and should not be construed as being limited to this
In the embodiment that illustrates, providing these embodiments on the contrary is in order to more thorough and be fully understood by the disclosure.It should be understood that
It is that being given for example only property of the accompanying drawings and embodiments effect of the disclosure is not intended to limit the protection scope of the disclosure.
It should be appreciated that each step recorded in disclosed method embodiment can execute in a different order,
And/or parallel execution.In addition, method implementation may include additional step and/or omit the step of execution is shown.This public affairs
The range opened is not limited in this respect.
Terms used herein " comprising " and its deformation are that opening includes, i.e., " including but not limited to ".Term "based"
It is " being based at least partially on ".Term " one embodiment " expression " at least one embodiment ";Term " another embodiment " indicates
" at least one other embodiment ";Term " some embodiments " expression " at least some embodiments ".The correlation of other terms is fixed
Justice provides in will be described below.
It is noted that the concepts such as " first " that refers in the disclosure, " second " are only used for different devices, module or list
Member distinguishes, and is not intended to limit the sequence or relation of interdependence of function performed by these devices, module or unit.
It is noted that referred in the disclosure "one", the modification of " multiple " be schematically and not restrictive this field
It will be appreciated by the skilled person that being otherwise construed as " one or more " unless clearly indicate otherwise in context.
The being merely to illustrate property of title of the message or information that are interacted between multiple devices in disclosure embodiment
Purpose, and be not used to limit the range of these message or information.
Embodiment one
Fig. 1 is the flow chart of one of the embodiment of the present disclosure one object detecting method, and the present embodiment is applicable to regarding
The case where target object is detected in frequency, this method can be executed by article detection device, the device can using software and/or
The mode of hardware realizes that the device can be configured in electronic equipment, such as mobile terminal, and typical mobile terminal includes hand
Machine, car-mounted terminal or laptop etc..As shown in Figure 1, this method specifically comprises the following steps:
S110 determines the target object region in initial video frame comprising examined object.
Initial video frame can refer to the video frame in video, specifically, video can refer to the video prerecorded, or
Person is the video of real-time recording.Correspondingly, initial video frame can be any one video frame in the video prerecorded, or
Person is the video frame obtained in real time during video record.
Object detection to video is really to carry out object detection to each video frame that video includes.
Optionally, examined object is planar object, in addition, examined object is the object of stationary state.It needs to illustrate
, stationary state refers to that the position of examined object in video is stablized constant.Illustratively, examined object be comprising
The picture of pattern.
Target object region is used to determine position range of the examined object in initial video frame, and is subsequent object
Detecting step reduces detection range.Specifically, what each video frame was made of a large amount of pixels, it can be established for video frame
Plane coordinate system, each pixel corresponding coordinate in the plane coordinate system.Target object region can be fixed shape
Region, such as rectangular area.Target object region can be determined by the coordinate of the boundary pixel point in region.For example, obtaining mesh
The coordinate for two pixels of mark object area determined to angular vertex determines target object region, as left upper apex and bottom right are pushed up
Point or bottom left vertex and right vertices.Wherein, target object region can refer to the part area intercepted from initial video frame
Domain.Target object region can not also be intercepted from initial video frame, in this regard, the embodiment of the present disclosure is not specifically limited.
In fact, in order to improve the speed of object detection, obtain target object region in initial video frame, at the same it is subsequent
Object detection is carried out in target object region, further to judge which partial region includes the object in target object region
Body causes detection data amount big, institute to avoid finely being detected using the image area of entire video frame as detection range
It needs resource more, detects slow-footed situation.
Target object region includes examined object.
Wherein, the detection method in target object region can there are many, such as machine learning model trained in advance or image
Matching algorithm etc. specifically can according to need the detection method of selection target object area, have in this regard, the embodiment of the present disclosure is not done
Body limitation.
Illustratively, the detection method of examined object can be image matching algorithm, for example, by current video frame according to
Preset rules are divided into multiple images block, wherein two image blocks can partly overlap.By the characteristic of examined object point
(for example, similarity calculation) is not compared with each image block, if each image block is mismatched with examined object, shows
The examined object is not present in the video frame;It is if it exists with the matched image block of examined object, matching degree is highest
Target object region of the image block as current video frame.Wherein, matching degree calculate specifically can be calculate image block with to
The number of matched characteristic point between detection object, or calculate the similarity between image block and examined object.Subsequent
In detection process, the target object region division that a upper video frame can be determined is multiple images block, repeats above-mentioned detection step
It is rapid to carry out target object region detection.Alternatively, at least one image block in the multiple images block that can also optionally divide carries out
It is detected in target object region, if it is determined that at least one image block is mismatched with examined object, then removes at least one
The region that image block except image block determines is as target object region;Otherwise make in the region at least one image block determined
For target object region.There are also other concrete modes to search for examined object region, has in this regard, the embodiment of the present disclosure is not done
Body limitation.
Specifically, examined object is picture, picture is usually the quadrangle for including pattern, and the characteristic of picture includes
The characteristic of quadrangle.At this point, it is also an option that edge detection, detects the straight line of quadrangle, by the area comprising quadrangle
Domain is as target object region.
In addition, image matching algorithm can according to need selection, matching algorithm, feature-based matching such as based on gray scale
Algorithm or the matching algorithm based on relationship etc., in this regard, the embodiment of the present disclosure is not specifically limited.
Optionally, the target object region in initial video frame comprising examined object is determined, comprising: in video record
In the process, whether detect in the first video frame in the video comprising examined object, if comprising, using first video frame as
Initial video frame;Otherwise, it continues to test whether comprising examined object in next video frame, until detecting comprising object to be detected
Body, the video frame comprising examined object is as initial video frame.
Specifically, object detection can be the detection process of real-time perfoming during video record.First video frame is used
In the beginning as object detection.First video frame can be first video frame of video record, be also possible in video record
During, when receiving object detection instruction, with the matched video frame of present system time.Wherein, object detection instructs
It can refer to the instruction that user is inputted by trigger action.
In a specific example, user can sweep control or other image recognition controls by touching to sweep, and call
Camera carries out the captured in real-time of video and the detection of examined object.
By the way that when confirmly detecting examined object, current video frame is determined as initial video frame, guarantee initial view
Include examined object in frequency frame, avoids the object for continuing subsequent successive video frames in the case where not including examined object
The wasting of resources and testing result mistake or invalid caused by detecting step, so that the resource utilization of object detection is improved,
And improve the accuracy rate of object detection.
S120, the traversal video from the initial video frame, according to the target object region in previous video frame, determination is worked as
Target object region in preceding video frame, until probability of the target object region comprising examined object in current video frame is big
In setting targets threshold.
After determining the target object region of initial video frame, in the latter video frame of initial video frame with the mesh
Mark object area carries out object detection in corresponding region, determines the target object region of latter video frame.Traverse video, circulation
Above-mentioned object detection step, until the probability comprising examined object in target object region is greater than setting targets threshold.
Optionally, the target object region according in previous video frame, determines the target object in current video frame
Region, comprising: according to region corresponding with the target object region of previous video frame in current video frame, determination is described to work as forward sight
The area to be tested of frequency frame;Examined object is detected in the area to be tested of the current video frame, obtains described working as forward sight
The target object region of frequency frame.
By replacing using the corresponding region in target object region of previous video frame as the range to be detected of current video frame
For current video frame all images area region as range to be detected, reduce the range of object detection, improve object inspection
The speed of survey.
Optionally, the object detecting method, further includes: if the target object region in current video frame is sky,
Detection is re-started using current video frame as new initial video frame.
It is understood that is, target object region is sky if can't detect target object in the video frame, show
It does not include target object in the video frame.At this point, being needed in subsequent video frame for the region in full images areal extent
Carry out target object identification.
By re-starting detection for current video frame as new initial video frame when target object region is empty,
To reduce the non-productive work of object detection, the accuracy rate of object detection is improved.
S130 determines object detection result according to the target object region of current video frame.
Object detection result is used to indicate the exact position range of target object.Probability comprising examined object is greater than
The target object region of targets threshold is set as object detection result.Alternatively, identifying mesh after determining target object region
The content for marking object area, using the content as object detection result.
Illustratively, examined object is picture, and target object region is picture region, and the content in target object region is
The content of picture, e.g., picture are trademark image, and the content of picture is the title of trade mark.
Optionally, until the probability that the target object region in current video includes examined object is greater than setting target
After threshold value, further includes: the determining target object Region Matching for being greater than setting targets threshold with the probability comprising examined object
Target video frame;The target object is shown in the video frame after the target video frame and the target video frame
Region.
Specifically, can be shown in current video frame and subsequent each video frame after determining object detection result
The matched target object detection zone of the object detection result.Illustratively, target object detection zone can pass through no filling
Wire mark display, target object detection zone can also have other display modes, in this regard, the embodiment of the present disclosure is not made specifically
Limitation.
By showing target object detection zone after determining object detection result, guarantee that the region of display is
Accurate target object detection zone.
The embodiment of the present disclosure is by beginning stepping through video from initial video frame, according to the target object region of previous video frame
The target object region for determining current video frame, reduces the detection range of each video frame, to reduce the testing number of each video frame
According to amount, while detection data amount reduces the raising that can bring detection speed;In addition, including object to be detected in target object region
When the probability of body is greater than setting targets threshold, determine object detection result, guarantee only determine target object region include to
When detection object, object detection result is just determined, to guarantee the accuracy rate of object detection, hereby it is achieved that taking into account target object
The speed and accuracy rate of detection solve and carry out detecting within the scope of full figure in the prior art to each video frame in video,
Lead to the low efficiency of object detection, the big problem of required operation resource is realized in the successive video frames passed through in video gradually
Precision target object area reduces the data volume of the image procossing of each video frame, improves the speed of target object detection, together
When, it takes into account and detects accurate target object region, to realize the speed and accuracy rate for taking into account object detection, and improve object
The efficiency that physical examination is surveyed.
Embodiment two
Fig. 2 a is the flow chart of one of embodiment of the present disclosure object detecting method, and this method is based on above-mentioned steps
It optimizes, the examined object is planar object.In step: according to the target object region of current video frame, determining object
Physical examination is surveyed as a result, being embodied as: carrying out plane geometry verification to the target object region of current video frame;According to by verification
Target object region determines object detection result.
This method specifically comprises the following steps:
S210 determines the target object region in initial video frame comprising examined object, and the examined object is flat
Face object.
Initial video frame, examined object and target object region in the embodiment of the present disclosure etc. can refer to above-mentioned implementation
The description of example.
Specifically, examined object is the picture comprising pattern, such as poster or trademark image etc..
S220, the traversal video from the initial video frame, according to the target object region in previous video frame, determination is worked as
Target object region in preceding video frame, until probability of the target object region comprising examined object in current video frame is big
In setting targets threshold.
S230 carries out plane geometry verification to according to the target object region of current video frame.
S240 determines object detection result according to the target object region by verifying.
Whether plane geometry verification is plane for verification object object area.When target object region passes through verification,
Show that the target object region is plane domain, to show that the target object is planar object.
Specifically, it is several to carry out plane to target object region by homography matrix (homography matrix)
What is verified.Further, it is also possible to which whether verifying target object region is planar object by other means, in this regard, the disclosure is implemented
Example is not particularly limited.
Optionally, it is operated not over the detection that the target object region of verification can continue to object area, or
Next video frame is re-started detection by person.
After determining that target object region is planar object, image knowledge further can be carried out to the target object region
Not, it is identified specifically, can according to need selection image-recognizing method, such as machine learning model or image trained in advance
Matching algorithm etc..In this regard, the embodiment of the present disclosure is not specifically limited.
Illustratively, the target object region can be identified by bag of words trained in advance.Specifically, extracting training
The feature of image, and the characteristic of extraction is handled, cluster forms multiple classes, and a type represents a visual word,
And the multiple class quantizations training image based on formation, obtain the code book of the training image.Meanwhile whole visual words are counted into shape
At dictionary.During target object region recognition, the characteristic point in target object region is extracted, and each characteristic point is traversed
Dictionary searches matched visual word.Using comprising the most corresponding image of code book of matched visual word number as target object
The recognition result in region.
Wherein, the dictionary that the code book of each training image and characteristic are formed can be depositing for preparatory training completion
It stores up in the terminal, alternatively, can also be by the subsequent progressive updating of server, and be issued to mobile terminal and stored.
In a specific example, as shown in Figure 2 b, examined object 244 is the picture comprising lightning pattern, is being worked as
It is detected in region 242 corresponding with the target object region of previous video frame in preceding video frame 241, obtains current video frame
241 target object region 243.
The embodiment of the present disclosure is by being greater than the probability comprising examined object and setting when examined object is planar object
Plane geometry verification is done in the target object region for the threshold value that sets the goal, and further judges whether target object region is plane, and
When determining that target object region is plane, object detection result is determined, realize and the feature in target object region is carried out further
Verifying improves object detection accuracy rate.
Embodiment three
Fig. 3 is a kind of structural schematic diagram for article detection device that the embodiment of the present disclosure three provides, and the present embodiment is applicable
In in video detect target object the case where.The device can realize by the way of software and/or hardware, which can be with
It is configured in electronic equipment, such as mobile terminal.As shown in figure 3, the apparatus may include: initial video frame object detection module
310, subsequent video frame object detection module 320 and object detection result determining module 330.
Initial video frame object detection module 310, for determining the object in initial video frame comprising examined object
Body region;
Subsequent video frame object detection module 320, for the traversal video from the initial video frame, according to previous video
Target object region in frame determines the target object region in current video frame, until the target object in current video frame
Region includes that the probability of examined object is greater than setting targets threshold;
Object detection result determining module 330 determines object detection for the target object region according to current video frame
As a result.
The embodiment of the present disclosure is by beginning stepping through video from initial video frame, according to the target object region of previous video frame
It determines the target object region of current video frame, realizes that a small range in current video frame carries out object detection, substitution is straight
Progress target object detection within the scope of the full figure in current video frame is connect, the data volume of small range detection is less than a wide range of detection
Data volume, to realize the detection data amount for reducing each video frame, while the reduction of detection data amount can bring detection speed
Raising;In addition, determining object detection when the probability that target object region includes examined object is greater than setting targets threshold
As a result, guaranteeing object detection result just to be determined, to guarantee object only when determining that target object region includes examined object
The accuracy rate that physical examination is surveyed solves in the prior art hereby it is achieved that taking into account the speed and accuracy rate of target object detection to video
In each video frame carry out detecting within the scope of full figure, lead to the low efficiency of object detection, the big problem of required operation resource,
It realizes through Query refinement target object region in successive video frames in video, reduces the image procossing of each video frame
Data volume improves the speed of target object detection, meanwhile, it takes into account and detects accurate target object region, taken into account to realize
The speed and accuracy rate of object detection, and improve the efficiency of object detection.
Optionally, the subsequent video frame object detection module 320, comprising: latter video frame object detection unit is used for
According to region corresponding with the target object region of previous video frame in current video frame, the to be checked of the current video frame is determined
Survey region;Examined object is detected in the area to be tested of the current video frame, obtains the target of the current video frame
Object area.
Optionally, the examined object is planar object.
Optionally, the object detection result determining module 330, comprising: plane geometry verification unit, for working as forward sight
The target object region of frequency frame carries out plane geometry verification;Object detection knot is determined according to the target object region by verifying
Fruit.
Optionally, the article detection device, further includes: detection module again, if for the mesh in current video frame
Marking object area is sky, then re-starts detection using current video frame as new initial video frame.
Optionally, the article detection device, further includes: target object region display module, for until working as forward sight
Target object region in frequency frame includes that the probability of examined object is greater than after setting targets threshold, determine with comprising to be detected
The probability of object is greater than the target video frame of the target object Region Matching of setting targets threshold;In the target video frame and
The target object region is shown in video frame after the target video frame.
Optionally, the initial video frame object detection module 310, comprising: initial video frame determination unit, for regarding
It during frequency is recorded, detects whether comprising examined object in the first video frame in the video, if comprising regarding head
Frequency frame is as initial video frame;Otherwise, it whether continues to test in next video frame comprising examined object, includes up to detecting
Examined object, the video frame comprising examined object is as initial video frame.
The article detection device that the embodiment of the present disclosure provides belongs to same with the object detecting method that previous embodiment provides
Inventive concept, the technical detail of detailed description not can be found in previous embodiment in the embodiments of the present disclosure, and the disclosure is implemented
The object detecting method beneficial effect having the same that the article detection device and previous embodiment that example provides provide.
Example IV
Below with reference to Fig. 4, it illustrates the electronic equipments (such as mobile terminal) for being suitable for being used to realize the embodiment of the present disclosure
400 structural schematic diagram.Terminal device in the embodiment of the present disclosure can include but is not limited to such as mobile phone, notebook electricity
Brain, digit broadcasting receiver, PDA (personal digital assistant), PAD (tablet computer), PMP (portable media player), vehicle
The mobile terminal of mounted terminal (such as vehicle mounted guidance terminal) etc. and such as number TV, desktop computer etc. are fixed eventually
End.Electronic equipment shown in Fig. 4 is only an example, should not function to the embodiment of the present disclosure and use scope bring it is any
Limitation.
As shown in figure 4, electronic equipment 400 may include processing unit (such as central processing unit, graphics processor etc.)
401, random access can be loaded into according to the program being stored in read-only memory (ROM) 402 or from storage device 408
Program in memory (RAM) 403 and execute various movements appropriate and processing.In RAM 403, it is also stored with electronic equipment
Various programs and data needed for 400 operations.Processing unit 401, ROM 402 and RAM 403 pass through the phase each other of bus 404
Even.Input/output (I/O) interface 405 is also connected to bus 404.
In general, following device can connect to I/O interface 405: including such as touch screen, touch tablet, keyboard, mouse, taking the photograph
As the input unit 406 of head, microphone, accelerometer, gyroscope etc.;Including such as liquid crystal display (LCD), loudspeaker, vibration
The output device 407 of dynamic device etc.;Storage device 408 including such as tape, hard disk etc.;And communication device 409.Communication device
409, which can permit electronic equipment 400, is wirelessly or non-wirelessly communicated with other equipment to exchange data.Although Fig. 4 shows tool
There is the electronic equipment 400 of various devices, it should be understood that being not required for implementing or having all devices shown.It can be with
Alternatively implement or have more or fewer devices.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising being carried on non-transient computer can
The computer program on medium is read, which includes the program code for method shown in execution flow chart.At this
In the embodiment of sample, which can be downloaded and installed from network by communication device 409, or be filled from storage
It sets 408 to be mounted, or is mounted from ROM 402.When the computer program is executed by processing unit 401, the disclosure is executed
The above-mentioned function of being limited in the method for embodiment.
Embodiment five
The above-mentioned computer-readable medium of the disclosure can be computer-readable signal media or computer-readable storage
Medium either the two any combination.Computer readable storage medium for example may be-but not limited to-electricity,
Magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Computer-readable storage
The more specific example of medium can include but is not limited to: have electrical connection, the portable computer magnetic of one or more conducting wires
Disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM or sudden strain of a muscle
Deposit), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or above-mentioned appoint
The suitable combination of meaning.In the disclosure, computer readable storage medium can be any tangible medium for including or store program,
The program can be commanded execution system, device or device use or in connection.And in the disclosure, computer
Readable signal medium may include in a base band or as the data-signal that carrier wave a part is propagated, wherein carrying computer
Readable program code.The data-signal of this propagation can take various forms, including but not limited to electromagnetic signal, optical signal
Or above-mentioned any appropriate combination.Computer-readable signal media can also be any other than computer readable storage medium
Computer-readable medium, the computer-readable signal media can be sent, propagated or transmitted for by instruction execution system, dress
It sets or device uses or program in connection.The program code for including on computer-readable medium can be with any
Medium transmission appropriate, including but not limited to: electric wire, optical cable, RF (radio frequency) etc. or above-mentioned any appropriate combination.
In some embodiments, client, server can use such as HTTP (HyperText Transfer
Protocol, hypertext transfer protocol) etc the network protocols of any currently known or following research and development communicated, and can
To be interconnected with the digital data communications (for example, communication network) of arbitrary form or medium.The example of communication network includes local area network
(" LAN "), wide area network (" WAN "), Internet (for example, internet) and ad-hoc network are (for example, the end-to-end net of ad hoc
Network) and any currently known or following research and development network.
Above-mentioned computer-readable medium can be included in above-mentioned electronic equipment;It is also possible to individualism, and not
It is fitted into the electronic equipment.
Above-mentioned computer-readable medium carries one or more program, when said one or multiple programs are by the electricity
When sub- equipment executes, so that the electronic equipment: determining the target object region in initial video frame comprising examined object;From institute
It states initial video frame traversal video and the target in current video frame is determined according to the target object region in previous video frame
Object area, until the probability that the target object region in current video frame includes examined object is greater than setting targets threshold;
According to the target object region of current video frame, object detection result is determined.
The calculating of the operation for executing the disclosure can be write with one or more programming languages or combinations thereof
Machine program code, above procedure design language include but is not limited to object oriented program language-such as Java,
Smalltalk, C++ further include conventional procedural programming language-such as " C " language or similar program design language
Speech.Program code can be executed fully on the user computer, partly be executed on the user computer, as an independence
Software package execute, part on the user computer part execute on the remote computer or completely in remote computer or
It is executed on server.In situations involving remote computers, remote computer can pass through the network of any kind --- packet
It includes local area network (LAN) or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as benefit
It is connected with ISP by internet).
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in module involved in the embodiment of the present disclosure can be realized by way of software, can also be by hard
The mode of part is realized.Wherein, the title of module does not constitute the restriction to the module itself under certain conditions, for example, just
Beginning video frame object detection module is also described as " determining the target object area in initial video frame comprising examined object
The module in domain ".
Function described herein can be executed at least partly by one or more hardware logic components.Example
Such as, without limitation, the hardware logic component for the exemplary type that can be used include: field programmable gate array (FPGA), specially
With integrated circuit (ASIC), Application Specific Standard Product (ASSP), system on chip (SOC), complex programmable logic equipment (CPLD) etc.
Deng.
In the context of the disclosure, machine readable media can be tangible medium, may include or is stored for
The program that instruction execution system, device or equipment are used or is used in combination with instruction execution system, device or equipment.Machine can
Reading medium can be machine-readable signal medium or machine-readable storage medium.Machine readable media can include but is not limited to electricity
Son, magnetic, optical, electromagnetism, infrared or semiconductor system, device or equipment or above content any conjunction
Suitable combination.The more specific example of machine readable storage medium will include the electrical connection of line based on one or more, portable meter
Calculation machine disk, hard disk, random access memory (RAM), read-only memory (ROM), Erasable Programmable Read Only Memory EPROM (EPROM
Or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage facilities or
Any appropriate combination of above content.
According to one or more other embodiments of the present disclosure, present disclose provides a kind of object detecting methods, apply in movement
In terminal, comprising:
Determine the target object region in initial video frame comprising examined object;
The traversal video from the initial video frame, according to the target object region in previous video frame, forward sight is worked as in determination
Target object region in frequency frame is set until the probability that the target object region in current video frame includes examined object is greater than
Set the goal threshold value;
According to the target object region of current video frame, object detection result is determined.
It is described according to previous in the object detecting method provided according to one or more other embodiments of the present disclosure, the disclosure
Target object region in video frame determines the target object region in current video frame, comprising: according in current video frame with
The corresponding region in target object region of previous video frame, determines the area to be tested of the current video frame;Described current
Examined object is detected in the area to be tested of video frame, obtains the target object region of the current video frame.
In the object detecting method provided according to one or more other embodiments of the present disclosure, the disclosure, the object to be detected
Body is planar object.
In the object detecting method provided according to one or more other embodiments of the present disclosure, the disclosure, the basis is current
The target object region of video frame, determines object detection result, comprising: carries out plane to the target object region of current video frame
Geometry verification;Object detection result is determined according to the target object region by verifying.
The object detecting method provided according to one or more other embodiments of the present disclosure, the disclosure, further includes: if current
Target object region in video frame is sky, then re-starts detection using current video frame as new initial video frame.
In the object detecting method provided according to one or more other embodiments of the present disclosure, the disclosure, until working as forward sight
Target object region in frequency frame includes that the probability of examined object is greater than after setting targets threshold, further includes: is determined and packet
Probability containing examined object is greater than the target video frame of the target object Region Matching of setting targets threshold;It is regarded in the target
The target object region is shown in video frame after frequency frame and the target video frame.
In the object detecting method provided according to one or more other embodiments of the present disclosure, the disclosure, initial video is determined
It include the target object region of examined object in frame, comprising: during video record, detect the first view in the video
It whether include examined object in frequency frame, if comprising using first video frame as initial video frame;Otherwise, it continues to test down
It whether include examined object in one video frame, until detecting the video frame comprising examined object, comprising examined object
As initial video frame.
According to one or more other embodiments of the present disclosure, present disclose provides a kind of article detection devices, are configured at movement
In terminal, comprising:
Initial video frame object detection module, for determining the target object area in initial video frame comprising examined object
Domain;
Subsequent video frame object detection module, for the traversal video from the initial video frame, according to previous video frame
In target object region, determine the target object region in current video frame, until current video frame in target object area
Domain includes that the probability of examined object is greater than setting targets threshold;
Object detection result determining module determines object detection knot for the target object region according to current video frame
Fruit.
In the article detection device provided according to one or more other embodiments of the present disclosure, the disclosure, the subsequent video
Frame object detection module, comprising: latter video frame object detection unit, for according in current video frame with previous video frame
The corresponding region in target object region, determines the area to be tested of the current video frame;In the to be checked of the current video frame
It surveys in region and detects examined object, obtain the target object region of the current video frame.
According to the article detection device that one or more other embodiments of the present disclosure, the disclosure provide, the examined object
For planar object.
In the article detection device provided according to one or more other embodiments of the present disclosure, the disclosure, the object detection
As a result determining module, comprising: plane geometry verification unit carries out plane geometry for the target object region to current video frame
Verification;Object detection result is determined according to the target object region by verifying.
In the article detection device provided according to one or more other embodiments of the present disclosure, the disclosure, the object detection
Device, further includes: detection module again, if being sky for the target object region in current video frame, with current video
Frame re-starts detection as new initial video frame.
According to the article detection device that one or more other embodiments of the present disclosure, the disclosure provide, the object detection dress
It sets, further includes: target object region display module, for until the target object region in current video frame includes to be detected
The probability of object is greater than after setting targets threshold, the determining mesh for being greater than setting targets threshold with the probability comprising examined object
Mark the matched target video frame of object area;It is shown in the video frame after the target video frame and the target video frame
Show the target object region.
According to the article detection device that one or more other embodiments of the present disclosure, the disclosure provide, the initial video frame
Object detection module, comprising: initial video frame determination unit, for detecting in the video during video record
It whether include examined object in first video frame, if comprising using first video frame as initial video frame;Otherwise, continue to examine
It surveys whether comprising examined object in next video frame, until detecting the view comprising examined object, comprising examined object
Frequency frame is as initial video frame.
According to one or more other embodiments of the present disclosure, present disclose provides a kind of electronic equipment, comprising:
One or more processors;
Memory, 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 realize the disclosure provide it is any as described in object detecting method.
According to one or more other embodiments of the present disclosure, present disclose provides a kind of computer readable storage mediums, thereon
Be stored with computer program, when which is executed by processor realize the disclosure provide it is any as described in object detecting method.
Above description is only the preferred embodiment of the disclosure and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that the open scope involved in the disclosure, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from design disclosed above, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed in the disclosure
Can technical characteristic replaced mutually and the technical solution that is formed.
Although this is not construed as requiring these operations with institute in addition, depicting each operation using certain order
The certain order that shows executes in sequential order to execute.Under certain environment, multitask and parallel processing may be advantageous
's.Similarly, although containing several specific implementation details in being discussed above, these are not construed as to this public affairs
The limitation for the range opened.Certain features described in the context of individual embodiment can also be realized in combination single real
It applies in example.On the contrary, the various features described in the context of single embodiment can also be individually or with any suitable
The mode of sub-portfolio is realized in various embodiments.
Although having used specific to this theme of the language description of structure feature and/or method logical action, answer
When understanding that theme defined in the appended claims is not necessarily limited to special characteristic described above or movement.On on the contrary,
Special characteristic described in face and movement are only to realize the exemplary forms of claims.
Claims (10)
1. a kind of object detecting method, which is characterized in that application is in the terminal, comprising:
Determine the target object region in initial video frame comprising examined object;
The traversal video from the initial video frame determines current video frame according to the target object region in previous video frame
In target object region, until current video frame in target object region include examined object probability be greater than setting mesh
Mark threshold value;
According to the target object region of current video frame, object detection result is determined.
2. the method according to claim 1, wherein the target object region according in previous video frame,
Determine the target object region in current video frame, comprising:
According to region corresponding with the target object region of previous video frame in current video frame, the current video frame is determined
Area to be tested;
Examined object is detected in the area to be tested of the current video frame, obtains the target object of the current video frame
Region.
3. the method according to claim 1, wherein the examined object is planar object.
4. according to the method described in claim 3, it is characterized in that, the target object region according to current video frame, really
Determine object detection result, comprising:
Plane geometry verification is carried out to the target object region of current video frame;
Object detection result is determined according to the target object region by verifying.
5. the method according to claim 1, wherein further include:
If target object region in current video frame is sky, using current video frame as new initial video frame again into
Row detection.
6. the method according to claim 1, wherein until the target object region in current video frame includes
The probability of examined object is greater than after setting targets threshold, further includes:
The target video frame of the determining target object Region Matching for being greater than setting targets threshold with the probability comprising examined object;
The target object region is shown in the video frame after the target video frame and the target video frame.
7. the method according to claim 1, wherein including examined object in the determining initial video frame
Target object region, comprising:
During video record, detect whether comprising examined object in the first video frame in the video, if comprising,
Then using first video frame as initial video frame;Otherwise, it continues to test whether comprising examined object in next video frame, until inspection
It measures comprising examined object, using the video frame comprising examined object as initial video frame.
8. a kind of article detection device, which is characterized in that be configured in mobile terminal, comprising:
Initial video frame object detection module, for determining the target object region in initial video frame comprising examined object;
Subsequent video frame object detection module, for the traversal video from the initial video frame, according in previous video frame
Target object region determines the target object region in current video frame, until the target object region packet in current video frame
Probability containing examined object is greater than setting targets threshold;
Object detection result determining module determines object detection result for the target object region according to current video frame.
9. a kind of electronic equipment characterized by comprising
One or more processors;
Memory, 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
Existing object detecting method as claimed in claim 1.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
Object detecting method as claimed in claim 1 is realized when execution.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910599369.1A CN110334650A (en) | 2019-07-04 | 2019-07-04 | Object detecting method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910599369.1A CN110334650A (en) | 2019-07-04 | 2019-07-04 | Object detecting method, device, electronic equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110334650A true CN110334650A (en) | 2019-10-15 |
Family
ID=68143109
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910599369.1A Pending CN110334650A (en) | 2019-07-04 | 2019-07-04 | Object detecting method, device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110334650A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110910415A (en) * | 2019-11-28 | 2020-03-24 | 重庆中星微人工智能芯片技术有限公司 | Parabolic detection method, device, server and computer readable medium |
CN112132024A (en) * | 2020-09-22 | 2020-12-25 | 中国农业大学 | Underwater target recognition network optimization method and device |
CN112258551A (en) * | 2020-03-18 | 2021-01-22 | 北京京东振世信息技术有限公司 | Article falling detection method, device, equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108038837A (en) * | 2017-12-08 | 2018-05-15 | 苏州科达科技股份有限公司 | Object detection method and system in video |
CN108564069A (en) * | 2018-05-04 | 2018-09-21 | 中国石油大学(华东) | A kind of industry safe wearing cap video detecting method |
CN108960090A (en) * | 2018-06-20 | 2018-12-07 | 腾讯科技(深圳)有限公司 | Method of video image processing and device, computer-readable medium and electronic equipment |
-
2019
- 2019-07-04 CN CN201910599369.1A patent/CN110334650A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108038837A (en) * | 2017-12-08 | 2018-05-15 | 苏州科达科技股份有限公司 | Object detection method and system in video |
CN108564069A (en) * | 2018-05-04 | 2018-09-21 | 中国石油大学(华东) | A kind of industry safe wearing cap video detecting method |
CN108960090A (en) * | 2018-06-20 | 2018-12-07 | 腾讯科技(深圳)有限公司 | Method of video image processing and device, computer-readable medium and electronic equipment |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110910415A (en) * | 2019-11-28 | 2020-03-24 | 重庆中星微人工智能芯片技术有限公司 | Parabolic detection method, device, server and computer readable medium |
CN112258551A (en) * | 2020-03-18 | 2021-01-22 | 北京京东振世信息技术有限公司 | Article falling detection method, device, equipment and storage medium |
CN112258551B (en) * | 2020-03-18 | 2023-09-05 | 北京京东振世信息技术有限公司 | Article drop detection method, device, equipment and storage medium |
CN112132024A (en) * | 2020-09-22 | 2020-12-25 | 中国农业大学 | Underwater target recognition network optimization method and device |
CN112132024B (en) * | 2020-09-22 | 2024-02-27 | 中国农业大学 | Underwater target recognition network optimization method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109214343B (en) | Method and device for generating face key point detection model | |
CN112184738B (en) | Image segmentation method, device, equipment and storage medium | |
CN109508681A (en) | The method and apparatus for generating human body critical point detection model | |
CN109584276A (en) | Critical point detection method, apparatus, equipment and readable medium | |
CN109447156B (en) | Method and apparatus for generating a model | |
CN109993150B (en) | Method and device for identifying age | |
EP3893125A1 (en) | Method and apparatus for searching video segment, device, medium and computer program product | |
TW202034270A (en) | Vehicle accident identification method and apparatus, and electronic device | |
CN110334650A (en) | Object detecting method, device, electronic equipment and storage medium | |
CN110335334A (en) | Avatars drive display methods, device, electronic equipment and storage medium | |
CN110413812A (en) | Training method, device, electronic equipment and the storage medium of neural network model | |
CN109919244A (en) | Method and apparatus for generating scene Recognition model | |
CN110532983A (en) | Method for processing video frequency, device, medium and equipment | |
CN109947989A (en) | Method and apparatus for handling video | |
CN110059623A (en) | Method and apparatus for generating information | |
CN110059624A (en) | Method and apparatus for detecting living body | |
CN111984803B (en) | Multimedia resource processing method and device, computer equipment and storage medium | |
CN109902446A (en) | Method and apparatus for generating information prediction model | |
CN110046571A (en) | The method and apparatus at age for identification | |
CN110223124A (en) | Data processing method, device, electronic equipment and computer-readable medium | |
CN109271929A (en) | Detection method and device | |
CN109829431B (en) | Method and apparatus for generating information | |
CN110414450A (en) | Keyword detection method, apparatus, storage medium and electronic equipment | |
CN111126159A (en) | Method, apparatus, electronic device, and medium for tracking pedestrian in real time | |
CN110287161A (en) | Image processing method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191015 |