CN110378934A - Subject detection method, apparatus, electronic equipment and computer readable storage medium - Google Patents
Subject detection method, apparatus, electronic equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN110378934A CN110378934A CN201910658738.XA CN201910658738A CN110378934A CN 110378934 A CN110378934 A CN 110378934A CN 201910658738 A CN201910658738 A CN 201910658738A CN 110378934 A CN110378934 A CN 110378934A
- Authority
- CN
- China
- Prior art keywords
- target image
- main body
- moving object
- target
- subject
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
This application involves a kind of subject detection method, apparatus, electronic equipment and computer readable storage mediums.The above method includes: to carry out motion detection to target image, determines the moving object and the corresponding movement velocity of moving object in target image；When movement velocity is more than threshold speed, subject detection is carried out to target image, obtains the candidate main body that target image includes；The target subject of target image is determined according to candidate main body and moving object.Since subject detection can be carried out to target image based on the movement velocity of moving object, so that the candidate main body and moving object that obtain according to detection determine the target subject of target image, the accuracy of subject detection can be improved.
Description
Technical field
This application involves image technology fields, more particularly to a kind of subject detection method, apparatus, electronic equipment and calculating
Machine readable storage medium storing program for executing.
Background technique
With the development of image technology, subject detection technology using more and more extensive.It is identified by subject detection technology
Main body in image can focus to main body, be tracked, the operation such as beauty Yan Meihua, Local treatment.Currently, subject detection skill
Art is based primarily upon deep learning algorithm to realize.However, traditional subject detection technology has that accuracy is low.
Summary of the invention
The embodiment of the present application provides a kind of subject detection method, apparatus, electronic equipment and computer readable storage medium, can
To improve the accuracy of subject detection.
A kind of subject detection method, comprising:
Motion detection is carried out to target image, determines that the moving object and the moving object in the target image correspond to
Movement velocity；
When the movement velocity is more than threshold speed, subject detection is carried out to the target image, obtains the target
The candidate main body that image includes；
The target subject of the target image is determined according to the candidate main body and the moving object.
A kind of subject detection device, comprising:
Motion detection block determines the moving object in the target image for carrying out motion detection to target image
Movement velocity corresponding with the moving object；
Subject detection module, for carrying out main body to the target image when the movement velocity is more than threshold speed
Detection, obtains the candidate main body that the target image includes；
Main body determining module, for determining the target of the target image according to the candidate main body and the moving object
Main body.
A kind of electronic equipment, including memory and processor store computer program, the calculating in the memory
When machine program is executed by the processor, so that the processor executes following steps:
Motion detection is carried out to target image, determines that the moving object and the moving object in the target image correspond to
Movement velocity；
When the movement velocity is more than threshold speed, subject detection is carried out to the target image, obtains the target
The candidate main body that image includes；
The target subject of the target image is determined according to the candidate main body and the moving object.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
Following steps are realized when row:
Motion detection is carried out to target image, determines that the moving object and the moving object in the target image correspond to
Movement velocity；
When the movement velocity is more than threshold speed, subject detection is carried out to the target image, obtains the target
The candidate main body that image includes；
The target subject of the target image is determined according to the candidate main body and the moving object.
Aforementioned body detection method, device, electronic equipment and computer readable storage medium can obtain in target image
Moving object and the corresponding movement velocity of moving object, when movement velocity be more than movement threshold when, detection target image include
Candidate main body subject detection can be improved to determine the target subject of target image according to moving object and candidate main body
Accuracy.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of application for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the structural block diagram of electronic equipment in one embodiment；
Fig. 2 is the flow chart of subject detection method in one embodiment；
Fig. 3 is the flow chart of the subject detection method provided in one embodiment；
Fig. 4 is the flow chart for rejecting moving object in one embodiment from candidate main body；
Fig. 5 is the flow chart for carrying out motion detection in one embodiment to target image；
Fig. 6 is the flow chart for carrying out subject detection in one embodiment to target image；
Fig. 7 is image processing effect schematic diagram in one embodiment；
Fig. 8 is the structural block diagram of subject detection device in one embodiment；
Fig. 9 is the schematic diagram of image processing circuit in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, and
It is not used in restriction the application.
It is appreciated that term " first " used in this application, " second " etc. can be used to describe various elements herein
And parameter, but these elements and parameter should not be limited by these terms.These terms are only used to by first element and another yuan
Part is distinguished, or by the first parameter and another parameter.It for example, can be in the case where not departing from scope of the present application
First segmentation figure is known as the second segmentation figure, and similarly, the second segmentation figure can be known as the first segmentation figure.First segmentation figure and
Second segmentation figure both segmentation figure, but it is not same segmentation figure.
Fig. 1 is the schematic diagram of internal structure of electronic equipment in one embodiment.As shown in Figure 1, the electronic equipment includes logical
Cross the processor and memory of system bus connection.Wherein, which supports entire electricity for providing calculating and control ability
The operation of sub- equipment.Memory may include non-volatile memory medium and built-in storage.Non-volatile memory medium is stored with behaviour
Make system and computer program.The computer program can be performed by processor, to be mentioned for realizing following each embodiment
A kind of subject detection method supplied.Built-in storage provides high speed for the operating system computer program in non-volatile memory medium
The running environment of caching.The electronic equipment can be mobile phone, tablet computer or personal digital assistant or wearable device etc..?
In some embodiments, which is also possible to server.Wherein, server can be independent server, be also possible to
It is realized by server cluster that multiple servers form.
Fig. 2 is the flow chart of subject detection method in one embodiment.As shown in Fig. 2, subject detection method includes step
202 to step 206, in which:
Step 202, motion detection is carried out to target image, determines that the moving object and moving object in target image correspond to
Movement velocity.
Target image can be the image that electronic equipment is acquired by camera, be also possible to be stored in electronic equipment local
Image or from network download image.Target image can also be video, the frame image in image sequence.Motion detection is
Refer to the operation for the moving object for including in identification image.Moving object refers to the object in image acquisition process there are change in location
Body.For example, common moving object includes pedestrian, animal, automobile etc..Movement velocity is used to indicate the change in location of moving object
Speed.In general, movement velocity can be calculated according to change in location size of the moving object in the image of imaging.
Electronic equipment carries out motion detection to target image, can determine the moving object and moving object in target image
Corresponding movement velocity.Specifically, the available at least two field pictures comprising target image of electronic equipment, thus using background
Relief method, frame differential method or optical flow method such as Lucas-Kanade Method (Lucas-card Nader method) etc. is to acquisition
At least two field pictures detected, then can determine moving object in target image and the corresponding movement speed of moving object
Degree.Optionally, the moving object that target image includes can be one or more, the corresponding movement of each moving object
Speed.
Step 204, when movement velocity is more than threshold speed, subject detection is carried out to target image, obtains target image
The candidate main body for including.
Electronic equipment can carry out main body inspection to target image when the movement velocity of moving object is more than threshold speed
It surveys, obtains the candidate main body that target image includes.Wherein, threshold speed can be set according to practical application request.Specifically,
By collecting image largely comprising moving object, the movement velocity of moving object and the shooting master of image in image are analyzed
Body can determine the threshold speed.
Specifically, when target image includes a moving object, then can be more than in the movement velocity of the moving object
When threshold speed, then subject detection is carried out to target image.When target image includes multiple moving objects, optionally, electronics
Equipment can obtain maximum movement velocity, when maximum fortune from the corresponding multiple movement velocitys of multiple moving objects
When dynamic speed is more than threshold speed, then subject detection is carried out to target image；Electronic equipment can also be in the maximum movement of area
When the movement velocity of object is more than threshold speed, then subject detection is carried out to target image；Moving object and figure can also be integrated
The distance of inconocenter is more than threshold speed in movement velocity minimum, the maximum moving object of area at a distance from picture centre
When, then subject detection etc. is carried out to target image, it is not limited here.
Electronic equipment carries out subject detection to target image, obtains the candidate main body that target image includes.Specifically, electronics
Equipment can carry out subject detection to target image by the subject detection model of deep learning.Electronic equipment can be by target figure
As being input to subject detection model, obtaining target image to target image progress subject detection by subject detection model includes
Multiple candidate main bodys.By candidate main body, corresponding pixel forms candidate main body in the picture.Specifically, when main body is examined
When surveying model and exporting the corresponding region of main body using body profile by the way of, then the edge pixel point of candidate main body is candidate master
The edge pixel point of the profile of body.Wherein, the candidate main body that target image includes is also possible to one or more.
Wherein, subject detection model can pass through deep learning algorithm such as CNN (Convolutional Neural
Network, convolutional neural networks), DNN (Deep Neural Network, deep neural network) or RNN (Recurrent
Neural Network, Recognition with Recurrent Neural Network) etc. realize.Optionally, in some embodiments, electronic equipment can prestore more
The corresponding image feature information of kind main body, by the image feature information of target image and the image feature information prestored progress
Match, then the corresponding main body of the image feature information of successful match is the candidate main body that the target image includes.
Step 206, the target subject of target image is determined according to candidate main body and moving object.
Target subject refers to the main body of finally determining target image.Electronic equipment can be right according to determining target subject
Target image optimizes processing, for example, electronic equipment can carry out U.S. face processing, color enhancement processing to target subject,
Virtualization processing can be carried out to target image according to target subject, camera pair can also be controlled under the scene of captured in real-time
Image Acquisition operation etc. is carried out after coke to target subject.
Electronic equipment determines the target subject of target image according to candidate main body and moving object.Specifically, electronic equipment
The area of candidate main body and moving object can be integrated, at a distance from picture centre, the movement velocity of moving object, candidate main body
At least one of confidence level etc. determine the target subject of target image.Wherein, the confidence level of candidate main body refers to target
It include the credibility of candidate's main body in image.Optionally, the available movement velocity of electronic equipment is no more than threshold speed
Moving object and area be more than area threshold target subject of the candidate main body as target image；Also available movement
Speed is no more than threshold speed and is detected as target subject of the moving object as target image of candidate main body simultaneously；Also
Moving object can be rejected from candidate main body, using the candidate main body after rejecting as target subject, or reject movement velocity
More than the moving object of threshold speed, using the candidate main body after rejecting as target subject etc., it is not limited here.
In embodiment provided by the present application, motion detection can be carried out to target image, determine the movement in target image
Object and the corresponding movement velocity of moving object then carry out main body inspection to target image when movement velocity is more than threshold speed
It surveys, obtains the candidate main body that target image includes, the target subject of target image is determined according to candidate main body and moving object.By
In can the movement velocity for the moving object that image includes be more than threshold speed when, according to the candidate main body and fortune of subject detection
Animal body determines the target subject of target image, and the accuracy of subject detection can be improved.
In one embodiment, electronic equipment can preset the corresponding target subject method of determination of different screening-modes,
It is true based on the target subject to obtain corresponding target subject method of determination according to the currently employed screening-mode of camera
Determine mode, the target subject of target image is determined according to candidate main body and moving object.For example, slow motion screening-mode is corresponding
Target subject method of determination can be the fortune that will acquire movement velocity no more than threshold speed and be detected as candidate main body simultaneously
Target subject of the animal body as target image；The corresponding target subject method of determination of portrait mode can be main from candidate
Moving object is rejected in body, using the candidate main body after rejecting as target subject；The corresponding target subject of screening-mode that jumps is true
Determine mode to can be movement velocity to be more than threshold speed, and is detected as the moving object of candidate main body simultaneously as target figure
The target subject etc. of picture.When screening-mode is that slow motion is shot, if it includes that movement velocity is small that motion detection, which obtains target image,
In the portrait A and portrait B of threshold speed, movement velocity is greater than the portrait C of threshold speed, and subject detection obtains target image and includes
When candidate main body is portrait A, portrait C and flower, then electronic equipment can be using portrait A as the target subject of target image.
Corresponding target subject method of determination is obtained by the screening-mode used according to camera, to be based on the target
Main body method of determination, the moving object of target image and candidate main body determine target subject, that is, distinguish different screening-modes and adopt
With different subject detection modes, the accuracy of subject detection can be improved.
In one embodiment, the subject detection method provided further include:, will when movement velocity is no more than threshold speed
Target subject of the moving object as target image.
Optionally, threshold speed is the maximum movement speed that moving object can be considered as shooting main body.That is, working as
When the movement velocity of moving object is more than the threshold speed, then the moving object can be determined not and be the main body of the target image；
And when the movement velocity of moving object is no more than the threshold speed, then the moving object can be determined as to the target image
In main body.
Electronic equipment can carry out motion detection to target image, obtain the moving object and moving object that target image includes
The corresponding movement velocity of body, when the movement velocity of moving object is no more than the threshold speed, then using the moving object as mesh
The target subject of logo image.Optionally, the movement velocity for the one or more moving objects for including when target image is no more than
When threshold speed, then the one or more moving object is determined as to the target subject of target image；When target image wraps simultaneously
It is more than the moving object of threshold speed and when no more than the moving object of threshold speed containing movement velocity, electronic equipment can integrate
Moving object is at a distance from picture centre, the size of moving object etc. determines the target movement for comparing with threshold speed
Object then determines the moving object that movement velocity is no more than threshold speed when target moving object is no more than threshold speed
For the target subject of target image.
In general, the movement velocity for shooting main body is usually smaller when carrying out image taking, movement velocity is biggish move it is past
Toward the background for being shooting, for example, there may be the biggish animals of movement velocity, the vehicle of traveling for background when carrying out portrait
Etc.；The embodiment of the present application is by carrying out motion detection to target image, in the movement speed for the moving object that target image includes
When degree is more than threshold speed, candidate and moving object that integration objective image includes determine the target subject of target image, are transporting
When the movement velocity of animal body is no more than threshold speed, then moving object is determined as to the target subject of target image, it can
The target subject that target image includes is detected based on the movement velocity of moving object, the accurate of subject detection can be improved
Property.
In one embodiment, the subject detection method provided can also include: when determining target image does not include movement
When object, subject detection is carried out to target image, obtains the candidate main body that target image includes；Using candidate main body as target figure
The target subject of picture.
There is the case where not including moving object in target image.Electronic equipment can carry out motion detection to target image
When, then determine that target image does not wrap when not exporting moving object that target image includes and the corresponding movement velocity of moving object
Containing moving object.Specifically, electronic equipment can analyze the multiple image comprising target image, when multiple image is corresponding
When image information is identical or difference is smaller, it is determined that do not include moving object in target image.
When target image does not include moving object, electronic equipment can carry out subject detection to target image, obtain mesh
The candidate main body that logo image includes, using candidate main body as the target subject of the target image.The candidate that subject detection obtains is main
Body can be one or more.Optionally, the candidate main body of one or more that electronic equipment can will test is used as mesh
The target subject of logo image can also choose the candidate main body as target subject from multiple candidate main bodys.Specifically, electronics is set
It is standby to obtain the corresponding classification of multiple candidate main bodys, the corresponding relationship based on classification and priority, by the class of highest priority
Not corresponding candidate main body is as target subject.For example, the priority of classification can be personage, animal, plant successively reduce,
It is not limited here.Optionally, it is big to can be combined with candidate main body position in the picture, the area of candidate main body for electronic equipment
It is one or more to determine target subject in confidence level of the corresponding classification of small, candidate main body etc..
In one embodiment, electronic equipment can adopt poolnet (detection method based on pond technology) subject detection
Mode carries out subject detection to target image, obtains the candidate main body that target image includes.Poolnet subject detection model introduces
GGM (Global Guidance Module, global guiding module) and FAM (Feature Aggregation Module, it is special
Sign integrates module), the details of the candidate main body of detection can be sharpened, and improve the efficiency of subject detection.
Fig. 3 is the flow chart of the subject detection method provided in one embodiment.As shown in figure 3, the subject detection method
Process it is as follows:
Step 302, motion detection is carried out to target image.
Step 304, judge whether target image includes moving object, if so, 306 are entered step, if it is not, then entering step
Rapid 316.
Step 306, moving object and the corresponding movement velocity of moving object that target image includes are obtained.
Step 308, judge whether movement velocity is more than threshold speed, if so, 310 are entered step, if it is not, then entering step
Rapid 314.
Step 310, subject detection is carried out to target image, obtains the candidate main body that target subject includes.
Step 312, the target subject of target image is determined according to candidate main body and moving object.
Step 314, using moving object as the target subject of target image.
Step 316, subject detection is carried out to target image, obtains the candidate main body that target subject includes.
Step 318, using candidate main body as the target subject of target image.
By to target image carry out motion detection, according to the result of motion detection use different main body methods of determination with
Determine the target subject of target image, subject detection result is more accurate, meets image taking demand.
In one embodiment, target image is determined according to candidate main body and moving object in the subject detection method provided
Target subject, comprising: the target image moving object that includes is rejected from candidate main body, using the candidate main body after rejecting as
Target subject.
Moving object is rejected from candidate main body, refers to and removes the main body of the moving object in candidate main body.Electronic equipment
The moving object that target image includes is rejected from candidate main body, specifically, electronic equipment can reject together from candidate main body
When be identified as the candidate main body of moving object, thus using the candidate main body after rejecting as target subject.For example, as shooting people
When picture, electronic equipment can carry out motion detection and obtain the moving object that target image includes to be portrait D, automobile to target image
E and automobile F；And the candidate main body that subject detection obtains includes portrait D, portrait E and automobile E, then electronic equipment can will be candidate
The portrait D and automobile E of main body are rejected, using candidate main body, that is, portrait E after rejecting as the target subject of target image.
When carrying out image taking, often there is some backgrounds, such as pedestrian, the vehicle of traveling etc., to image
When carrying out subject detection, these backgrounds can also be identified as main body.By the way that in target image, there are movement velocitys to be greater than speed threshold
The moving object of value carries out subject detection to target image, the candidate main body for including in target image is obtained, from candidate main body
Moving object is rejected, to obtain the target subject of target image, the accuracy of subject detection is higher.
Fig. 4 is the flow chart for rejecting moving object in one embodiment from candidate main body.As shown in figure 4, in a reality
It applies in example, the moving object that target image includes is rejected in the subject detection method provided from candidate main body, after rejecting
Process of the candidate main body as target subject, comprising:
Step 402, corresponding first segmentation figure of candidate main body is obtained, and obtains corresponding second segmentation figure of moving object.
First segmentation figure is output when carrying out subject detection to target image.First segmentation figure identifies candidate main body in mesh
Position in logo image.First segmentation figure can be binarization segmentation figure, and the candidate in target image is respectively indicated using 0 and 1
Other regions except main body and candidate main body.For example, the first segmentation figure can be the corresponding picture of candidate main body of target image
The pixel value of vegetarian refreshments is 1, and the pixel value of the pixel in addition to the corresponding pixel of candidate main body is 0.Similar, the second segmentation
Figure is output when carrying out motion detection to target image.Second segmentation figure identifies the position of moving object in the target image
It sets.Second segmentation figure is also possible to binarization segmentation figure.It is worth noting that, in this embodiment, it is candidate in the first segmentation figure
The pixel of moving object uses identical binary value in the pixel of main body and the second segmentation figure, it can all for 1 or
It is all 0.In some other implementation, in the first segmentation figure in the pixel of candidate main body and the second segmentation figure moving object picture
Vegetarian refreshments can be different using identical binary value.
Step 404, intermediate segmentation figure is generated based on the first segmentation figure and the second segmentation figure；Intermediate segmentation figure includes first point
Cut the overlapping region of figure and the second segmentation figure.
Intermediate segmentation figure includes the overlapping region of the first segmentation figure and the second segmentation figure.Specifically, in this embodiment,
The pixel of moving object uses identical binary value in the pixel of candidate main body and the second segmentation figure in one segmentation figure, then
Available first segmentation figure of electronic equipment is identical with position in the second segmentation figure and the identical pixel of pixel value, by these
Pixel constitutes intermediate segmentation figure.Optionally, electronic equipment takes the intersection of the first segmentation figure Yu the second segmented image, in obtaining
Between segmentation figure.The first segmentation figure is indicated with M, and N indicates the second segmentation figure, then intermediate segmentation figure I=M ∩ N.
Step 406, according to the first segmentation figure and intermediate segmentation figure, target subject segmentation figure is determined.
Target subject segmentation figure identifies the position of target subject in the target image.Electronic equipment is according to target subject point
Target subject in target image can be determined by cutting figure.First segmentation figure is the corresponding segmentation figure of candidate main body.Electronic equipment root
Target subject segmentation figure is determined according to the first segmentation figure and intermediate segmentation figure, and specifically, electronic equipment is by the first segmentation figure and intermediate
Segmentation figure subtracts operation, i.e., rejects moving object from candidate main body.According to above-mentioned example, the first segmentation figure is indicated with M, N is indicated
Second segmentation figure, intermediate segmentation figure I=M ∩ N, then target subject divides J=M-J=M-M ∩ N.With the first segmentation figure and second
In segmentation figure using 1 as candidate main body and moving object corresponding pixel points pixel value for be illustrated, obtained target
The pixel value of the corresponding pixel of target subject is 1 in main body segmentation figure, and its in addition to the corresponding pixel of target subject
The pixel value of his pixel is 0, and electronic equipment can determine the target subject in target image according to the target subject segmentation figure.
By obtaining corresponding first segmentation figure of candidate main body and corresponding second segmentation figure of moving object, thus based on the
One segmentation figure and the second segmentation figure obtain target subject segmentation figure, and the accuracy of target subject can be improved.
In one embodiment, the movement that target image includes is provided from candidate main body in the subject detection method provided
Object, using the candidate main body after rejecting as the process of target subject, comprising: rejecting target image from candidate main body includes
Moving object；By in each candidate main body after rejecting, area is more than the candidate main body of area threshold as target subject.
Area threshold can obtain after the corresponding size of main body is for statistical analysis by great amount of images,
This is without limitation.For example, area threshold can be 30%, 40%, 50% of image area etc..Optionally, area threshold can also
To be determined according to the area of each candidate main body after the quantity of required target subject and rejecting.For example, after rejecting
The sizes of 4 candidate main bodys be respectively 300*300,320*320,400*400,450*450, target subject if desired
Quantity be 3, then area threshold can be 350*350 or 360*360 etc..Optionally, the area of candidate main body can also use
The ratio of the area of the area and target image of candidate main body indicates.
In one embodiment, electronic equipment can also be by each candidate main body after rejecting, the maximum candidate of area
Main body is as target subject.
By rejecting the target image moving object that includes from candidate main body, by each candidate main body after rejecting,
Area be more than area threshold candidate main body as target subject, it can to target subject progress postsearch screening, by area compared with
Big candidate main body can be further improved the accuracy of target subject as target subject.
Fig. 5 is the flow chart for carrying out motion detection in one embodiment to target image.As shown in figure 5, the main body provided
Motion detection is carried out to target image in detection method, determines the moving object and the corresponding movement of moving object in target image
The process of speed, comprising:
Step 502, the image sequence comprising target image is obtained.
Target image can be any one frame image in image sequence.Optionally, target image can be located at video or
Middle position or position to the rear in person's image sequence, it can pass through preposition frame image and before and after frames image detection target figure
As comprising moving object.Image sequence can be video, and the multiframe preview image for being also possible to camera real-time capture is formed
Image sequence.Electronic equipment can carry out motion detection to target image according to the image sequence comprising target image.It is optional
Ground, the corresponding exposure parameter of the multiple image that image sequence includes are identical.
Step 504, position of the analysis pixel in each frame image that image sequence includes.
In the two field pictures being imaged respectively before and after moving object campaign, the corresponding pixel value of moving object should be kept not
Become.It is also possible to, the fortune in target image can be analyzed by the identical pixel of pixel value in detection before and after frames image
Animal body.
Electronic equipment analyzes position of the pixel in each frame image that image sequence includes.Specifically, electronic equipment
It can analyze position of the identical pixel of pixel value in each frame image.
Step 506, the pixel that change in location will be present is determined as moving pixel, and obtains in target image by moving
The moving object of pixel composition.
If there is variation in position of the identical pixel of pixel value in different images, then it is assumed that the pixel belongs to movement
The pixel is determined as moving pixel by the pixel of object, electronic equipment.Electronic equipment obtains in target image by moving
The moving object of pixel composition.Specifically, the connected region that electronic equipment can be made of movement pixel, then each connects
The logical corresponding moving object in region.Optionally, electronic equipment can filter the connected region that pixel quantity is less than amount threshold
Filtered connected region is determined as the corresponding region of moving object by domain.
In one embodiment, the change in location of pixel can be determined according to position of the pixel in each frame image
Amplitude；When change in location amplitude is more than change threshold, it is determined that pixel is movement pixel.
Change in location amplitude refers to the amplitude of change in location of the identical same pixel of pixel value in different images.
Electronic equipment can determine the change in location amplitude of pixel according to position of the pixel in each frame image.For example, to scheme
The center of picture be origin establish XY reference axis, if position of the pixel in each frame image be respectively (2000,3122),
(2000,3124), (1998,3122), (2001,3126), then the pixel is 2 pixels in the change in location amplitude of X-direction,
Change in location amplitude in the Y direction is 6 pixels.
Change threshold can be set according to practical application request, it is not limited here.Specifically, change threshold can pass through
It analyzes and shakes or shoot in the multiple image shot when object shake in camera, the change in location of the identical pixel of pixel value
Amplitude determines.For example, change threshold can be 4 pixels, 5 pixels, 6 pixels, 7 pictures using pixel as measurement unit
Element etc., it is not limited here.Optionally, electronic equipment can be set corresponding to the change threshold of X-direction and corresponding to Y-direction
Change threshold.Change in location amplitude is more than change threshold by the corresponding change in location amplitude of the available pixel of electronic equipment
Pixel be determined as move pixel, to obtain the moving object being made of in target image movement pixel.
In image shoot process, being easy to appear camera shake or shooting main body leads to shooting there are slight jitter
Multiple image has that the position of the identical pixel of pixel value in the picture is inconsistent.By the position for obtaining pixel
The pixel that change in location amplitude is more than change threshold is determined as moving pixel, to obtain target image by amplitude of variation
In the moving object that is made of movement pixel, can shake and shoot pixel caused by main body shake to avoid due to camera
The inconsistent problem in the position of point in the picture, can be improved the accuracy in motion detection.
Step 508, position of the movement pixel for including according to moving object in each frame image calculates moving object
Movement velocity.
Based in the two field pictures being imaged respectively before and after moving object campaign, the corresponding pixel value of moving object is remained unchanged
Principle, can determine I1 (x, y, t-1)=I2 (x+u, y+v, t), I1 and I2 indicate pixel respectively in image 1 and image 2
In position, t indicate the unit time, in this embodiment, t be acquire image 1 and image 2 time interval, u be the direction x on
The movement speed of pixel, v are the movement speed of pixel on the direction y.Electronic equipment can be according to above-mentioned formula, by moving object
The movement pixel that body includes substitutes into formula in the position in each frame image, has both obtained the movement of each movement pixel
Speed, average value, median or the mode of movement speed etc. by seeking each movement pixel can determine moving object
The movement velocity of body.
By position of the analysis pixel in each frame image that image sequence includes, obtain by there are change in location
The moving object of pixel composition is moved, position of the movement pixel for including according to moving object in each frame image calculates
The movement velocity of moving object, the then corresponding movement velocity of moving object and moving object that available target image includes.
In one embodiment, electronic equipment can carry out motion detection to target image by frame differential method, specifically
Ground, electronic equipment can subtract each other the pixel value of the pixel of two field pictures corresponding position adjacent in image sequence, obtained figure
As in, if pixel value is 0, it is determined that the pixel belongs to the pixel of stationary object or background, if pixel value is not 0,
Illustrate that the pixel is the corresponding pixel of moving object.Motion detection is carried out to target image by frame differential method, is calculated
Simply, the efficiency of motion detection can be improved.
Fig. 6 is the flow chart for carrying out subject detection in one embodiment to target image.As shown in fig. 6, the main body provided
Subject detection is carried out to target image in detection method, obtains the process for the candidate main body that target image includes, comprising:
Step 602, center weight figure corresponding with target image is generated, wherein weighted value represented by center weight figure
It is gradually reduced from center to edge.
Wherein, center weight figure refers to the figure for recording the weighted value of each pixel in target image.Center weight
The weighted value recorded in figure is gradually reduced from center to four sides, i.e., center weight is maximum, is gradually reduced again to four side rights.In
The weighted value of picture centre pixel to the image edge pixels point of heart weight map characterization target image is gradually reduced.
Electronic equipment can generate corresponding center weight figure according to the size of target image.Represented by the center weight figure
Weighted value be gradually reduced from center to four sides.Center weight figure can be used Gaussian function or use first-order equation or second order side
Cheng Shengcheng.The Gaussian function can be two-dimensional Gaussian function.
Step 604, target image and center weight figure are input in subject detection model, obtain body region confidence level
Figure.
Subject detection model be the visible light figure previously according to Same Scene, depth map, center weight figure and it is corresponding
The model that the main body exposure mask figure of mark is trained.Specifically, subject detection model is a large amount of training number of preparatory acquisition
According to being input to training data includes that the subject detection model of initial network weight is trained.Every group of trained number
According to including the corresponding visible light figure of Same Scene, center weight figure and the main body exposure mask figure marked.Wherein, it is seen that light figure is in
Input of the heart weight map as the subject detection model of training, main body exposure mask (mask) figure marked are examined as the main body of training
Survey the true value (ground truth) that model desired output obtains.Main body exposure mask figure is the image of main body in image for identification
Filter template can filter out the main body in image with the other parts of shielded image.Subject detection model can training can identify
Detect various main bodys, such as people, flower, cat, dog, background.
Specifically, target image and center weight figure can be input in subject detection model by electronic equipment, be detected
Available body region confidence level figure.Body region confidence level figure is to belong to the main body which kind of can be identified for recording main body
Probability, such as it is 0.8 that some pixel, which belongs to the probability of people, colored probability is 0.1, and the probability of background is 0.1.
Step 606, the candidate main body in target image is determined according to body region confidence level figure.
Wherein, main body refers to various objects, such as people, flower, cat, dog, ox, blue sky, white clouds, vehicle.
Specifically, electronic equipment can be chosen according to body region confidence level figure confidence level be greater than one of confidence threshold value or
Multiple main bodys are as candidate main body.Confidence threshold value can be set according to practical application request, it is not limited here.
In one embodiment, electronic equipment can be handled the body region confidence level figure, obtain main body exposure mask
Figure, detects the highlight area in the target image, and according to the highlight area and the main body exposure mask figure in the target image, determining should
The candidate main body of bloom is eliminated in target image.Wherein, it is lower, scattered that there are some confidence levels in body region confidence level figure
Point, electronic equipment is by being filtered the available main body exposure mask figure of processing to body region confidence level figure.The filtration treatment can
Using configuration confidence threshold value, the pixel by confidence value in body region confidence level figure lower than confidence threshold value is filtered.It should
Self-adapting confidence degree threshold value can be used in confidence threshold value, can also use fixed threshold, can also be corresponding using subregion configuration of territory
Threshold value.Highlight area refers to that brightness value is greater than the region of luminance threshold.Electronic equipment can carry out bloom inspection to target image
It surveys, screening obtains the target pixel points that brightness value is greater than luminance threshold, obtains bloom using Connected area disposal$ to target pixel points
Region, thus, the highlight area in target image is done into Difference Calculation with the main body exposure mask figure or target is calculated in logical AND
The candidate main body of bloom is eliminated in image.
Optionally, electronic equipment can also carry out at self-adapting confidence degree threshold filtering the body region confidence level figure
Reason, obtains binaryzation exposure mask figure；Morphological scale-space and guiding filtering processing are carried out to the binaryzation exposure mask figure, obtain main body exposure mask
Figure.Specifically, after electronic equipment handles body region confidence level figure according to self-adapting confidence degree threshold filtering, by the picture of reservation
The confidence value of vegetarian refreshments indicates that the confidence value of the pixel removed is indicated using 0 using 1, obtains binaryzation exposure mask figure.Form
Processing may include corrosion and expansion.Etching operation first can be carried out to binaryzation exposure mask figure, then carry out expansive working, removal is made an uproar
Sound；Filtering processing is guided to the binaryzation exposure mask figure after Morphological scale-space again, edge filter operation is realized, obtains to edge and mention
The main body exposure mask figure taken.The noise of main body exposure mask figure that can be guaranteed by Morphological scale-space and guiding filtering processing it is few or
There is no noise, edge is softer.
In one embodiment, what electronic equipment can also obtain that target image includes according to body region confidence level figure is more
The size of a main body and corresponding classification, priority level and main body corresponding region based on the corresponding classification of each main body determines
Candidate main body.Optionally, electronic equipment can preset different classes of corresponding priority level.For example, the priority level of classification can
To be that people, flower, cat, dog, ox, white clouds successively reduce.Priority level of the electronic equipment based on the corresponding classification of each main body and
The size in region is determined as candidate main body, specifically, when there is the multiple main bodys for belonging to the same category in target image, electricity
The maximum preset quantity main body in region can be determined as candidate main body according to the corresponding area size of multiple main bodys by sub- equipment；
When there is the multiple main bodys to belong to a different category in target image, electronic equipment can be corresponding by the highest classification of priority level
Main body as candidate main body, can be further according to more if there are the highest multiple main bodys of priority level in target image
The size of a main body region determines candidate main body.Optionally, electronic equipment can be combined with each main body in the picture
Position determine candidate main body.For example, electronic equipment can also preset different classes of priority level, different zones size, with
And the score value of the different location of main body in the picture, according to the priority level of the corresponding classification of each main body, region
Size, position in the picture calculate the fractional value of each main body, using the highest preset quantity main body of fractional value as time
Select main body.
The object of picture centre can be allowed to be easier to be detected by center weight figure, utilize visible light using trained
The subject detection model that the training such as figure, center weight figure and main body exposure mask figure obtains, can more accurately identify target figure
Candidate main body as in.
Fig. 7 is image processing effect schematic diagram in one embodiment.As shown in fig. 7, there are a butterflies in target image 702
Butterfly obtains body region confidence level Figure 70 6 after target image 702 to be input to the network 704 of subject detection model, then to master
Body region confidence level Figure 70 6 is filtered and obtains binaryzation exposure mask Figure 70 8 with binaryzation, then carries out to binaryzation exposure mask Figure 70 8
Morphological scale-space and guiding filtering realize edge enhancing, obtain main body exposure mask Figure 71 0.Body mask Figure 71 0 identifies candidate
Main body, that is, butterfly position in the target image.
It should be understood that although each step in the flow chart of Fig. 2-6 is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 2-6
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
Fig. 8 is the structural block diagram of subject detection device in one embodiment.As shown in figure 8, in one embodiment, main body
Detection device includes:
Motion detection block 802, for carrying out motion detection to target image, determine moving object in target image and
The corresponding movement velocity of moving object.
Subject detection module 804, for carrying out subject detection to target image when movement velocity is more than threshold speed,
Obtain the candidate main body that target image includes.
Main body determining module 806, for determining the target subject of target image according to candidate main body and moving object.
Embodiment provided by the present application can carry out motion detection to target image, determine the moving object in target image
Body and the corresponding movement velocity of moving object then carry out subject detection to target image when movement velocity is more than threshold speed,
The candidate main body that target image includes is obtained, the target subject of target image is determined according to candidate main body and moving object.Due to
It can be when the movement velocity for the moving object that image includes be more than threshold speed, according to the candidate main body of subject detection and movement
Object determines the target subject of target image, and the accuracy of subject detection can be improved.
In one embodiment, main body determining module 806 can be also used for when movement velocity is no more than threshold speed, will
Target subject of the moving object as target image.
In one embodiment, subject detection module 804 can be also used for not including moving object when determining target image
When, subject detection is carried out to target image, obtains the candidate main body that target image includes；Main body determining module 806 will be for that will wait
Select target subject of the main body as target image.
In one embodiment, main body determining module 806 can be also used for from candidate main body reject target image include
Moving object, using the candidate main body after rejecting as target subject.
In one embodiment, the beneficial effect that main body determining module 806 can be also used for Installation practice obtains candidate
Corresponding first segmentation figure of main body, and obtain corresponding second segmentation figure of moving object；Divided based on the first segmentation figure and second
Figure generates intermediate segmentation figure；Intermediate segmentation figure includes the overlapping region of the first segmentation figure and the second segmentation figure；According to the first segmentation
Figure and intermediate segmentation figure, determine target subject segmentation figure.
In one embodiment, main body determining module 806 can be also used for from candidate main body reject target image include
Moving object；By in each candidate main body after rejecting, area is more than the candidate main body of area threshold as target subject.
In one embodiment, motion detection block 802 can be also used for obtaining the image sequence comprising target image；Point
Position of the exploring vegetarian refreshments in each frame image that image sequence includes；The pixel that change in location will be present is determined as motor image
Vegetarian refreshments, and obtain the moving object being made of in target image movement pixel；The movement pixel for including according to moving object
Position in each frame image calculates the movement velocity of moving object.
In one embodiment, motion detection block 802 can be also used for the position according to pixel in each frame image
Set the change in location amplitude of determining pixel；When change in location amplitude is more than change threshold, it is determined that pixel is motor image
Vegetarian refreshments.
In one embodiment, subject detection module 804 can be also used for generating center weight corresponding with target image
Figure, wherein weighted value represented by center weight figure is gradually reduced from center to edge；Target image and center weight figure is defeated
Enter into subject detection model, obtains body region confidence level figure；It is determined in target image according to body region confidence level figure
Candidate main body.
The division of modules is only used for for example, in other embodiments in aforementioned body detection device, can be by master
Body detection device is divided into different modules as required, to complete all or part of function of aforementioned body detection device.
Realizing for the modules in subject detection device provided in the embodiment of the present application can be the shape of computer program
Formula.The computer program can be run on an electronic device.The program module that the computer program is constituted is storable in electronic equipment
Memory on.When the computer program is executed by processor, realize the embodiment of the present application described in method the step of.
The embodiment of the present application also provides a kind of electronic equipment.It include image processing circuit in above-mentioned electronic equipment, at image
Reason circuit can use hardware and or software component realization, it may include define ISP (Image Signal Processing, figure
As signal processing) the various processing units of pipeline.Fig. 9 is the schematic diagram of image processing circuit in one embodiment.Such as Fig. 9 institute
Show, for purposes of illustration only, only showing the various aspects of image processing techniques relevant to the embodiment of the present application.
As shown in figure 9, image processing circuit includes ISP processor 940 and control logic device 950.Imaging device 910 captures
Image data handled first by ISP processor 940, ISP processor 940 to image data analyzed with capture can be used for really
The image statistics of fixed and/or imaging device 910 one or more control parameters.Imaging device 910 may include having one
The camera of a or multiple lens 912 and imaging sensor 914.Imaging sensor 914 may include colour filter array (such as
Bayer filter), imaging sensor 914 can obtain the luminous intensity captured with each imaging pixel of imaging sensor 914 and wavelength
Information, and the one group of raw image data that can be handled by ISP processor 940 is provided.Sensor 920 (such as gyroscope) can be based on biography
The parameter (such as stabilization parameter) of the image procossing of acquisition is supplied to ISP processor 940 by 920 interface type of sensor.Sensor 920
Interface can use SMIA (Standard Mobile Imaging Architecture, Standard Mobile Imager framework) interface,
The combination of other serial or parallel camera interfaces or above-mentioned interface.
In addition, raw image data can also be sent to sensor 920 by imaging sensor 914, sensor 920 can be based on biography
Raw image data is supplied to ISP processor 940 to 920 interface type of sensor or sensor 920 deposits raw image data
It stores up in video memory 930.
ISP processor 940 handles raw image data pixel by pixel in various formats.For example, each image pixel can
Bit depth with 8,10,12 or 14 bits, ISP processor 940 can carry out raw image data at one or more images
Reason operation, statistical information of the collection about image data.Wherein, image processing operations can be by identical or different bit depth precision
It carries out.
ISP processor 940 can also receive image data from video memory 930.For example, 920 interface of sensor will be original
Image data is sent to video memory 930, and the raw image data in video memory 930 is available to ISP processor 940
It is for processing.Video memory 930 can be independent special in a part, storage equipment or electronic equipment of memory device
It with memory, and may include DMA (Direct Memory Access, direct direct memory access (DMA)) feature.
When receiving from 914 interface of imaging sensor or from 920 interface of sensor or from video memory 930
When raw image data, ISP processor 940 can carry out one or more image processing operations, such as time-domain filtering.Treated schemes
As data can be transmitted to video memory 930, to carry out other processing before shown.ISP processor 940 is from image
Memory 930 receives processing data, and carries out in original domain and in RGB and YCbCr color space to the processing data
Image real time transfer.Treated that image data may be output to display 970 for ISP processor 940, for user's viewing and/or
It is further processed by graphics engine or GPU (Graphics Processing Unit, graphics processor).In addition, ISP processor
940 output also can be transmitted to video memory 930, and display 970 can read image data from video memory 930.?
In one embodiment, video memory 930 can be configured to realize one or more frame buffers.In addition, ISP processor 940
Output can be transmitted to encoder/decoder 960, so as to encoding/decoding image data.The image data of coding can be saved,
And it is decompressed before being shown in 970 equipment of display.Encoder/decoder 960 can be real by CPU or GPU or coprocessor
It is existing.
The statistical data that ISP processor 940 determines, which can be transmitted, gives control logic device Unit 950.For example, statistical data can wrap
Include the image sensings such as automatic exposure, automatic white balance, automatic focusing, flicker detection, black level compensation, 912 shadow correction of lens
914 statistical information of device.Control logic device 950 may include the processor and/or micro-control for executing one or more routines (such as firmware)
Device processed, one or more routines can statistical data based on the received, determine the control parameter and ISP processor of imaging device 910
940 control parameter.For example, the control parameter of imaging device 910 may include 920 control parameter of sensor (such as gain, exposure
The time of integration, stabilization parameter of control etc.), camera flash control parameter, 912 control parameter of lens (such as focus or zoom
With focal length) or these parameters combination.ISP control parameter may include for automatic white balance and color adjustment (for example, in RGB
During processing) 912 shadow correction parameter of gain level and color correction matrix and lens.
In embodiment provided by the present application, imaging device 910 can be used for acquiring target image, video memory 930
It can be used for the target image of the acquisition of storage imaging equipment 910 and the image sequence comprising target image.ISP processor 940 can be with
Motion detection is carried out to target image, the corresponding movement velocity of moving object in target image is determined, when movement velocity is more than
When threshold speed, subject detection is carried out to target image, the candidate main body that target image includes is obtained, according to candidate main body and fortune
Animal body determines the target subject of target image.To, control logic device 950 can focus to the target subject, U.S. face,
The processing such as beautification.Subject detection side provided by above-described embodiment may be implemented by above-mentioned image processing circuit in electronic equipment
Method, details are not described herein.
The embodiment of the present application also provides a kind of computer readable storage mediums.One or more is executable comprising computer
The non-volatile computer readable storage medium storing program for executing of instruction, when the computer executable instructions are executed by one or more processors
When, so that the step of processor executing subject detection method.
A kind of computer program product comprising instruction, when run on a computer, so that computer executing subject
Detection method.
It may include non-to any reference of memory, storage, database or other media used in the embodiment of the present application
Volatibility and/or volatile memory.Suitable nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM), it is used as external cache.By way of illustration and not limitation, RAM in a variety of forms may be used
, such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDR SDRAM),
Enhanced SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM).
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
The limitation to the application the scope of the patents therefore cannot be interpreted as.It should be pointed out that for those of ordinary skill in the art
For, without departing from the concept of this application, various modifications and improvements can be made, these belong to the guarantor of the application
Protect range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (12)
1. a kind of subject detection method characterized by comprising
Motion detection is carried out to target image, determines the moving object and the corresponding fortune of the moving object in the target image
Dynamic speed；
When the movement velocity is more than threshold speed, subject detection is carried out to the target image, obtains the target image
The candidate main body for including；
The target subject of the target image is determined according to the candidate main body and the moving object.
2. the method according to claim 1, wherein the method also includes:
When the movement velocity is no more than the threshold speed, using the moving object as the target master of the target image
Body.
3. the method according to claim 1, wherein the method also includes:
When determining that the target image does not include moving object, subject detection is carried out to the target image, obtains the mesh
The candidate main body that logo image includes；
Using the candidate main body as the target subject of the target image.
4. the method according to claim 1, wherein described true according to the candidate main body and the moving object
The target subject of the fixed target image, comprising:
The moving object that the target image includes is rejected from the candidate main body, using the candidate main body after rejecting as described in
Target subject.
5. according to the method described in claim 4, it is characterized in that, described reject the target image from the candidate main body
The moving object for including, using the candidate main body after rejecting as the target subject, comprising:
Corresponding first segmentation figure of the candidate main body is obtained, and obtains corresponding second segmentation figure of the moving object；
Intermediate segmentation figure is generated based on first segmentation figure and the second segmentation figure；The intermediate segmentation figure includes described first point
Cut the overlapping region of figure and the second segmentation figure；
According to first segmentation figure and the intermediate segmentation figure, target subject segmentation figure is determined.
6. according to the method described in claim 4, it is characterized in that, described reject the target image from the candidate main body
The moving object for including, using the candidate main body after rejecting as the target subject, comprising:
The moving object that the target image includes is rejected from the candidate main body；
By in each candidate main body after rejecting, area is more than the candidate main body of area threshold as the target subject.
7. the method according to claim 1, wherein it is described to target image carry out motion detection, determine described in
Moving object and the corresponding movement velocity of the moving object in target image, comprising:
Obtain the image sequence comprising the target image；
Analyze position of the pixel in each frame image that described image sequence includes；
The pixel that change in location will be present is determined as moving pixel, and obtains in the target image by movement pixel group
At moving object；
Position of the movement pixel for including according to the moving object in each frame image calculates the fortune of the moving object
Dynamic speed.
8. the method according to the description of claim 7 is characterized in that the pixel that change in location will be present is determined as moving
Pixel, comprising:
The change in location amplitude of the pixel is determined according to position of the pixel in each frame image；
When the change in location amplitude is more than change threshold, it is determined that the pixel is movement pixel.
9. method according to any one of claim 1 to 8, which is characterized in that described to be led to the target image
Physical examination is surveyed, and the candidate main body that the target image includes is obtained, comprising:
Generate center weight figure corresponding with the target image, wherein weighted value represented by the center weight figure is therefrom
The heart is gradually reduced to edge；
The target image and the center weight figure are input in subject detection model, body region confidence level figure is obtained；
The candidate main body in the target image is determined according to the body region confidence level figure.
10. a kind of subject detection device characterized by comprising
Motion detection block determines the moving object in the target image and institute for carrying out motion detection to target image
State the corresponding movement velocity of moving object；
Subject detection module, for carrying out subject detection to the target image when the movement velocity is more than threshold speed,
Obtain the candidate main body that the target image includes；
Main body determining module, for determining the target master of the target image according to the candidate main body and the moving object
Body.
11. a kind of electronic equipment, including memory and processor, computer program, the calculating are stored in the memory
When machine program is executed by the processor, so that the processor executes main body inspection as claimed in any one of claims 1-9 wherein
The step of survey method.
12. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method as claimed in any one of claims 1-9 wherein is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910658738.XA CN110378934B (en) | 2019-07-22 | 2019-07-22 | Subject detection method, apparatus, electronic device, and computer-readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910658738.XA CN110378934B (en) | 2019-07-22 | 2019-07-22 | Subject detection method, apparatus, electronic device, and computer-readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110378934A true CN110378934A (en) | 2019-10-25 |
| CN110378934B CN110378934B (en) | 2021-09-07 |
Family
ID=68254559
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910658738.XA Active CN110378934B (en) | 2019-07-22 | 2019-07-22 | Subject detection method, apparatus, electronic device, and computer-readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110378934B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110866486A (en) * | 2019-11-12 | 2020-03-06 | Oppo广东移动通信有限公司 | Subject detection method and apparatus, electronic device, and computer-readable storage medium |
| CN111557692A (en) * | 2020-04-26 | 2020-08-21 | 深圳华声医疗技术股份有限公司 | Automatic measurement method, ultrasonic measurement device and medium for target organ tissue |
| CN113766130A (en) * | 2021-09-13 | 2021-12-07 | 维沃移动通信有限公司 | Video shooting method, electronic equipment and device |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103561271A (en) * | 2013-11-19 | 2014-02-05 | 福建师范大学 | Video airspace tamper detection method for removing moving object shot by static camera lens |
| CN103905726A (en) * | 2012-12-27 | 2014-07-02 | 佳能株式会社 | Imaging apparatus and method for controlling the same |
| US20160182792A1 (en) * | 2013-09-27 | 2016-06-23 | Fujifilm Corporation | Imaging device and imaging method |
| CN105827952A (en) * | 2016-02-01 | 2016-08-03 | 维沃移动通信有限公司 | Photographing method for removing specified object and mobile terminal |
| CN106997589A (en) * | 2017-04-12 | 2017-08-01 | 上海联影医疗科技有限公司 | image processing method, device and equipment |
| CN108347563A (en) * | 2018-02-07 | 2018-07-31 | 广东欧珀移动通信有限公司 | Method for processing video frequency and device, electronic equipment, computer readable storage medium |
| CN108712609A (en) * | 2018-05-17 | 2018-10-26 | Oppo广东移动通信有限公司 | Focusing process method, apparatus, equipment and storage medium |
| CN109167910A (en) * | 2018-08-31 | 2019-01-08 | 努比亚技术有限公司 | focusing method, mobile terminal and computer readable storage medium |
| CN109389135A (en) * | 2017-08-03 | 2019-02-26 | 杭州海康威视数字技术股份有限公司 | A kind of method for screening images and device |
-
2019
- 2019-07-22 CN CN201910658738.XA patent/CN110378934B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103905726A (en) * | 2012-12-27 | 2014-07-02 | 佳能株式会社 | Imaging apparatus and method for controlling the same |
| US20160182792A1 (en) * | 2013-09-27 | 2016-06-23 | Fujifilm Corporation | Imaging device and imaging method |
| CN103561271A (en) * | 2013-11-19 | 2014-02-05 | 福建师范大学 | Video airspace tamper detection method for removing moving object shot by static camera lens |
| CN105827952A (en) * | 2016-02-01 | 2016-08-03 | 维沃移动通信有限公司 | Photographing method for removing specified object and mobile terminal |
| CN106997589A (en) * | 2017-04-12 | 2017-08-01 | 上海联影医疗科技有限公司 | image processing method, device and equipment |
| CN109389135A (en) * | 2017-08-03 | 2019-02-26 | 杭州海康威视数字技术股份有限公司 | A kind of method for screening images and device |
| CN108347563A (en) * | 2018-02-07 | 2018-07-31 | 广东欧珀移动通信有限公司 | Method for processing video frequency and device, electronic equipment, computer readable storage medium |
| CN108712609A (en) * | 2018-05-17 | 2018-10-26 | Oppo广东移动通信有限公司 | Focusing process method, apparatus, equipment and storage medium |
| CN109167910A (en) * | 2018-08-31 | 2019-01-08 | 努比亚技术有限公司 | focusing method, mobile terminal and computer readable storage medium |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110866486A (en) * | 2019-11-12 | 2020-03-06 | Oppo广东移动通信有限公司 | Subject detection method and apparatus, electronic device, and computer-readable storage medium |
| CN110866486B (en) * | 2019-11-12 | 2022-06-10 | Oppo广东移动通信有限公司 | Subject detection method and apparatus, electronic device, and computer-readable storage medium |
| US12039767B2 (en) | 2019-11-12 | 2024-07-16 | Guangdong Oppo Mobile Telecommunications Corp., Ltd | Subject detection method and apparatus, electronic device, and computer-readable storage medium |
| CN111557692A (en) * | 2020-04-26 | 2020-08-21 | 深圳华声医疗技术股份有限公司 | Automatic measurement method, ultrasonic measurement device and medium for target organ tissue |
| CN113766130A (en) * | 2021-09-13 | 2021-12-07 | 维沃移动通信有限公司 | Video shooting method, electronic equipment and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110378934B (en) | 2021-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110334635B (en) | Subject tracking method, apparatus, electronic device and computer-readable storage medium | |
| US11457138B2 (en) | Method and device for image processing, method for training object detection model | |
| CN110149482B (en) | Focusing method, focusing device, electronic equipment and computer readable storage medium | |
| CN110248096B (en) | Focusing method and device, electronic equipment and computer readable storage medium | |
| WO2021022983A1 (en) | Image processing method and apparatus, electronic device and computer-readable storage medium | |
| CN113766125B (en) | Focusing method and device, electronic equipment and computer readable storage medium | |
| CN110248097B (en) | Focus tracking method and device, terminal equipment and computer readable storage medium | |
| US11538175B2 (en) | Method and apparatus for detecting subject, electronic device, and computer readable storage medium | |
| CN110191287B (en) | Focusing method and device, electronic equipment and computer readable storage medium | |
| CN108810413A (en) | Image processing method and device, electronic equipment, computer readable storage medium | |
| CN109712177B (en) | Image processing method, image processing device, electronic equipment and computer readable storage medium | |
| CN107911625A (en) | Light measuring method, device, readable storage medium storing program for executing and computer equipment | |
| CN110490196B (en) | Subject detection method and apparatus, electronic device, and computer-readable storage medium | |
| CN110378934A (en) | Subject detection method, apparatus, electronic equipment and computer readable storage medium | |
| CN110276831B (en) | Method and device for constructing three-dimensional model, equipment and computer-readable storage medium | |
| CN109327626A (en) | Image-pickup method, device, electronic equipment and computer readable storage medium | |
| CN110399823B (en) | Subject tracking method and apparatus, electronic device, and computer-readable storage medium | |
| CN112581481B (en) | Image processing method and device, electronic equipment and computer readable storage medium | |
| CN110688926B (en) | Subject detection method and apparatus, electronic device, and computer-readable storage medium | |
| CN107145820B (en) | Binocular positioning method based on HOG characteristics and FAST algorithm | |
| CN110689007B (en) | Subject recognition method and device, electronic equipment and computer-readable storage medium | |
| CN108881740A (en) | Image method and device, electronic equipment, computer readable storage medium | |
| CN110545384B (en) | Focusing method and device, electronic equipment and computer readable storage medium | |
| CN113784026A (en) | Method, apparatus, device and storage medium for calculating position information based on image | |
| CN110610171A (en) | Image processing method and device, electronic equipment and computer readable 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |