CN110334736A - Image-recognizing method, device, electronic equipment and medium - Google Patents
Image-recognizing method, device, electronic equipment and medium Download PDFInfo
- Publication number
- CN110334736A CN110334736A CN201910475041.9A CN201910475041A CN110334736A CN 110334736 A CN110334736 A CN 110334736A CN 201910475041 A CN201910475041 A CN 201910475041A CN 110334736 A CN110334736 A CN 110334736A
- Authority
- CN
- China
- Prior art keywords
- target object
- identification
- identification window
- image
- application
- 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
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (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
This application discloses a kind of image-recognizing method, device, electronic equipment and media.Wherein, in the application, after obtaining target object type, the available identification window to match with target object type, and image is calculated using identification window.By the technical solution of application the application, it can use the identification window to match with the type of target object and deep learning calculating carried out to complete the identification to target object to image.And then can to avoid image recognition it is incorrect caused by influence depth study calculated result the drawbacks of.
Description
Technical field
Image processing techniques involved in the application, especially a kind of image-recognizing method, device, electronic equipment and medium.
Background technique
Due to the rise of communication era and society, deep learning algorithm is continuous with the use of more and more users
Development.
Wherein, the main application scenarios of deep learning algorithm are the object for obtaining the user institute interest in multiple images
Body, so as to know the classification and specific location of target object.In computer vision, this generic task is known as mesh by us
Mark detection (object detection) or object detection.Further, the figure for needing to obtain shooting in some specific areas
Target object in piece is labeled to complete the detection to target using identification window, such as: in automatic Pilot field, need
Vehicle, pedestrian, tricycle, bicycle, electric vehicle, motorcycle etc. are labeled.It is needed in fields such as e- Learnings
Teacher or student etc. are labeled.
However, being usually present when identifying the target object in image using deep learning algorithm due to selected knowledge
The problem that other window is inaccurate and causes recognition efficiency not high.
Summary of the invention
The embodiment of the present invention provides a kind of image-recognizing method, device, electronic equipment and medium.
Wherein, according to the one aspect of the embodiment of the present application, a kind of image-recognizing method for providing, which is characterized in that packet
It includes:
Determine target object type, the target object type is for reflecting in the picture, the classification of the target object；
Based on the target object type, identification window, the identification window and the target object type phase are determined
Match；
Based on the identification window, the identification of the target object is carried out to described image.
It is optionally, described to be based on the target object type in another embodiment based on the application above method,
Determine identification window, comprising:
Based on the target object type, the Aspect Ratio of the target object is determined；
Based on the Aspect Ratio, the Aspect Ratio of the identification window is determined.
Optionally, described to be based on the Aspect Ratio in another embodiment based on the application above method, it determines
The Aspect Ratio of identification window, comprising:
Determine shared regional percentage of the target object in the images to be recognized；
Based on the shared regional percentage, the area of the identification window is calculated；
Area and the Aspect Ratio based on the identification window, obtain the identification window.
Optionally, in another embodiment based on the application above method, when the object in described image
When body quantity is multiple, based on the type of the maximum target object of area in the target object, identification window is determined.
Optionally, described to be based on the identification window in another embodiment based on the application above method, to institute
It states image and carries out identifying processing, comprising:
Described image is intercepted based on the identification window, obtains at least one subgraph, to it is described at least one
Subgraph carries out the identification of the target object.
Optionally, described to be based on the identification window to institute in another embodiment based on the application above method
It states image to be intercepted, obtains at least one subgraph, comprising:
The identification window is slided in described image by horizontal direction and/or vertical direction, successively intercepts, obtains institute
State at least one subgraph；And/or
By horizontal direction and/or vertical direction, multiple identification windows are arranged simultaneously in described image, are cut respectively
Obtain at least one described subgraph.
Optionally, described to be based on the identification window in another embodiment based on the application above method, to institute
State the identification that image carries out the target object, comprising:
Based on trained neural network image semantic segmentation model and the identification window in advance, in described image
The target object is identified.
According to the other side of the embodiment of the present application, a kind of pattern recognition device for providing, comprising:
Module is obtained, is configured to determine that target object type, the target object type is for reflecting in the picture, institute
State the classification of target object；
Determining module, is configured as based on the target object type, determines identification window, the identification window with it is described
Target object type matches；
Identification module is configured as carrying out described image the identification of the target object based on the identification window.
According to the another aspect of the embodiment of the present application, a kind of electronic equipment that provides, comprising:
Memory, for storing executable instruction；And
Display, for being shown with the memory to execute the executable instruction to complete any of the above-described figure
As the operation of recognition methods.
According to the still another aspect of the embodiment of the present application, a kind of computer readable storage medium provided, based on storing
The instruction that calculation machine can be read, described instruction are performed the operation for executing any of the above-described described image recognition methods.
In the application, after obtaining target object type, the available recognition window to match with target object type
Mouthful, and image is calculated using identification window.By the technical solution of application the application, can use and target object
The identification window that type matches carries out deep learning to image and calculates to complete the identification to target object.And then it can be to avoid
Image recognition it is incorrect caused by influence depth learn calculated result the drawbacks of.
Below by drawings and examples, the technical solution of the application is described in further detail.
Detailed description of the invention
The attached drawing for constituting part of specification describes embodiments herein, and together with description for explaining
The principle of the application.
The application can be more clearly understood according to following detailed description referring to attached drawing, in which:
Fig. 1 is the system architecture schematic diagram of the application image-recognizing method.
Fig. 2 is the mark schematic diagram of identification window in the picture.
Fig. 3 is the flow chart of another embodiment of the application image-recognizing method.
Fig. 4 a- Fig. 4 c is the schematic diagram of the application image-recognizing method.
Fig. 5 is the flow chart of another embodiment of the application image-recognizing method.
Fig. 6 a- Fig. 6 b is the schematic diagram of the application image-recognizing method.
Fig. 7 is the structural schematic diagram of the application pattern recognition device.
Fig. 8 is that the application shows electronic devices structure schematic diagram.
Specific embodiment
The various exemplary embodiments of the application are described in detail now with reference to attached drawing.It should also be noted that unless in addition having
Body explanation, the unlimited system of component and the positioned opposite of step, numerical expression and the numerical value otherwise illustrated in these embodiments is originally
The range of application.
Simultaneously, it should be appreciated that for ease of description, the size of various pieces shown in attached drawing is not according to reality
Proportionate relationship draw.
Be to the description only actually of at least one exemplary embodiment below it is illustrative, not as to the application and
Its any restrictions applied or used.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable
In the case of, the technology, method and apparatus should be considered as part of specification.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, then in subsequent attached drawing does not need that it is further discussed.
It is to be appreciated that the directional instruction (such as up, down, left, right, before and after ...) of institute is only used in the embodiment of the present application
In explaining in relative positional relationship, the motion conditions etc. under a certain particular pose (as shown in the picture) between each component, if should
When particular pose changes, then directionality instruction also correspondingly changes correspondingly.
In addition, the description for being such as related to " first ", " second " in this application is used for description purposes only, and should not be understood as
Its relative importance of indication or suggestion or the quantity for implicitly indicating indicated technical characteristic.Define as a result, " first ",
The feature of " second " can explicitly or implicitly include at least one of the features.In the description of the present application, " multiples' " contains
Justice is at least two, such as two, three etc., unless otherwise specifically defined.
In this application unless specifically defined or limited otherwise, term " connection ", " fixation " etc. shall be understood in a broad sense,
For example, " fixation " may be a fixed connection, it may be a detachable connection, or integral；It can be mechanical connection, be also possible to
Electrical connection；It can be directly connected, the connection inside two elements or two can also be can be indirectly connected through an intermediary
The interaction relationship of a element, unless otherwise restricted clearly.It for the ordinary skill in the art, can basis
Concrete condition understands the concrete meaning of above-mentioned term in this application.
It in addition, the technical solution between each embodiment of the application can be combined with each other, but must be general with this field
Based on logical technical staff can be realized, it will be understood that when the combination of technical solution appearance is conflicting or cannot achieve this
The combination of technical solution is not present, also not this application claims protection scope within.
It is described below with reference to Fig. 1 to Fig. 6 according to the application illustrative embodiments for carrying out image-recognizing method.
It should be noted that following application scenarios are merely for convenience of understanding spirit herein and principle and showing, the reality of the application
It is unrestricted in this regard to apply mode.On the contrary, presently filed embodiment can be applied to applicable any scene.
The application also proposes a kind of image-recognizing method, device, target terminal and medium.
Fig. 1 schematically shows a kind of flow diagram of image-recognizing method according to the application embodiment.Such as
Shown in Fig. 1, this method comprises:
S101 obtains target object type, and target object type is for reflecting in the picture, the classification of target object.
Firstly the need of explanation, in the application, the equipment for obtaining target object type is not specifically limited, for example,
The application can obtain target object type by smart machine, can also obtain target object type by server.
In addition, not being specifically limited to smart machine in the application, i.e., smart machine can be arbitrary smart machine,
For example, mobile phone, electronic memo, PDA etc..
Wherein, the target object type in the application is for reflecting in the picture, the classification of target object.For example, with net
For network online education field, when tester want identification image in teacher hand motion with judge its give lessons efficiency when, mesh
Mark object is the hand of teacher.Further, target object type is hand limbs.Likewise, when tester wants to know
The face action of other image middle school student with judge its listen to the teacher state when, target object is the face of student.Further, target
Object type is face organ.
Optionally, the target object type referred in the application can also be based on other objects, for example, classroom, blackboard, vapour
Vehicle etc., the application do not limit this.
S102 is based on target object type, determines that identification window, identification window match with target object type.
The application can also further be based on the target object type after obtaining target object type, obtain with
Matched identification window.
Firstly, the application is not specifically limited identification window, such as identification window can be bounding box.Utilizing depth
In the target detection that learning algorithm is realized, we describe target position usually using bounding box (bounding box).Wherein,
Bounding box is a rectangle frame, can be determined by the x and y-axis coordinate in the rectangle upper left corner and the x and y-axis coordinate in the lower right corner.User
Target object in image can be defined according to the coordinate information in image.As shown in Fig. 2, in e- Learning, always
Frame image when Shi Jinhang gives lessons.When tester wants the hand motion by teacher in neural network recognization image to judge
Its give lessons efficiency when, can by identification window load in the hand limbs of teacher, to realize identification window by target object
Overall profile carry out frame choosing so that the preset neural network model of later use, the object chosen automatically to identification window
Body is calculated, to achieve the purpose that automatic identification target object.
Further, in the related art, during carrying out automatic identification to target object using identification window, usually
There is a problem of leading to calculated result inaccuracy because the selection of identification window is improper.In the application, target is being got
After the target object type of object, the identification window to match with the type can be chosen according to the target object type.
It should be noted that in the application after getting target object type, can be obtained therewith from database
The identification window matched.Matching target identification window can also be automatically generated according to the target object type.The application couple
This is without limitation.
Further alternative, the application is not specifically limited identification window equally, i.e. identification window can be and target
The identification window that object type size matches, identification window may be the recognition window to match with target object type area
Mouthful.The application does not limit this.
S103 is based on identification window, and the identification of target object is carried out to image.
Optionally, in the application after getting the identification window to match with target object type, can be based on should
Identification window realizes the identification to target object.
In the application, after obtaining target object type, the available recognition window to match with target object type
Mouthful, and image is calculated using identification window.By the technical solution of application the application, can use and target object
The identification window that type matches carries out deep learning to image and calculates to complete the identification to target object.And then it can be to avoid
Image recognition it is incorrect caused by influence depth learn calculated result the drawbacks of.
It is further alternative, in a kind of embodiment of the application, in S102 (it is based on target object type, obtains mesh
Mark identification window) in, it further include a kind of specific embodiment, as shown in Figure 3, comprising:
S201 obtains target object type.
S202 is based on target object type, determines the Aspect Ratio of target object.
It, can be according to the target object type, really after obtaining the corresponding target object type of target object in the application
Set the goal the Aspect Ratio of object.Wherein, the Aspect Ratio of target object is the length of the object and the length ratio of width.Example
It such as, is 120cm for a length, width is for the blackboard of 40cm, and Aspect Ratio is 3:1.
In a kind of possible embodiment of the application, mesh can be determined according to target object type and preset strategy
Mark the Aspect Ratio of object.For example, when target object type be user hand limbs when, can according to inquiry preset strategy,
The Aspect Ratio for determining corresponding hand is 2:1.It, can be pre- according to inquiry when target object type is the face organ of user
If tactful, determine that the Aspect Ratio of corresponding face organ is 1.5:1.Again or, when target object type is the blackboard in classroom,
It can determine that the Aspect Ratio of corresponding face organ is 3:1 according to inquiry preset strategy.
It should be noted that not being specifically limited to preset strategy in the application, i.e., each mesh for including in preset strategy
The corresponding Aspect Ratio of mark object type can constantly be acquired according to developer to be calculated and corresponding change.Preset strategy it is specific
Variation will not influence the protection scope of the application.
S203 is based on Aspect Ratio, determines the Aspect Ratio of identification window.
Optionally, it in the application after determining the Aspect Ratio of target object, can further obtain with the length and width
The target identification window of ratio.For example, determining corresponding length according to preset strategy when target object is the hand limbs of user
Wide ratio is 2:1.Further, the target identification window that Aspect Ratio is similarly 2:1 can be obtained according to the Aspect Ratio.
It should be noted that not being specifically limited to the mode for obtaining identification window in the application.For example, the application can be with
After determining the Aspect Ratio of target object, corresponding identification window is obtained from database.The application can also determine mesh
After the Aspect Ratio for marking object, the identification window of corresponding Aspect Ratio is automatically generated.
In a kind of preferred embodiment of the application, it can also be accomplished by the following way based on Aspect Ratio, determine
Identification window:
Determine the shared regional percentage of target object in the picture；
Further, after determining the Aspect Ratio of target object, the suitable identification window of selection is ensured that, also
Need further to determine the size of target object.It should be understood that as shown in fig. 4 a, it can be seen that contain two in image
The face of user, and face organ's size of two users is different.Although namely the target object class of two user's face
Type is identical (can choose the identification window with identical aspect ratio example), but not due to the size of two target objects
Together, so if two identical identification windows of selection are easy to appear the biggish target object of wherein area and can not be identified window
Cover complete problem.Further, it only can just be selected after the ratio and size for determining target object for it
The target identification window of target object is framed completely.
Optionally, it in the embodiment for determining the size of target object in the picture, needs to obtain target first
The shared regional percentage of object in the picture.For example, then determining object when target object occupies the region of image half
The shared regional percentage of body in the picture is 50%.Likewise, when target object occupies the region of image 1/4th, then
Determine that the shared regional percentage of target object in the picture is 25%.
Based on shared regional percentage, the area of target identification window is determined；
Further, due to the size of every image be substantially it is fixed, the application is determining that target object scheming
After shared regional percentage as in, the size of target object can be determined according to the shared regional percentage.Further
, so that subsequent during choosing identification window, selection can cover the identification window of the target object.
Area and Aspect Ratio based on target identification window obtain target identification window.
In another embodiment of the application, obtain target object shared regional percentage in the picture it
Before, it can also be realized by following steps:
When the target object quantity in image is multiple, the class based on the maximum target object of area in target object
Type determines identification window.
Further, realize that the mode of above-mentioned steps can be such that
Obtain image；
In detection image, the quantity of target object；
When the quantity for detecting target object is multiple, it is big that the area of multiple target objects in the picture is obtained respectively
It is small；
By in multiple target objects, the maximum target object of area is as target object.
Optionally, when there are when multiple target objects, utilizing deep learning algorithm to carry out image to improve in image
The efficiency of identification can select the maximum target object of area as target object in the application in multiple target objects.With
Avoid the occurrence of target object it is too small caused by recognition result inaccuracy drawback.
By taking Fig. 4 b as an example, it can be seen that there are the hand limbs of multiple users in image.Further, such as Fig. 4 c institute
Show, when target object is hand limbs, the application can obtain simultaneously hand limbs A, B of multiple user, C, D, E, F,
G, and further in the hand limbs of multiple users, select the maximum hand limbs A of area as target object.
S204 is based on identification window, and the identification of target object is carried out to image.
Optionally, identification window is based in the application, identifying to image can be accomplished by the following way:
Based on trained in advance neural network image semantic segmentation model and identification window, image is identified.
Further, realize that the process of above-mentioned steps can be with are as follows:
Obtain sample image, wherein sample image includes the sample characteristics of at least one target object.
Preset neural network image semantic segmentation model is trained using sample image, obtains meeting preset condition
Neural network image semantic segmentation model.
Optionally, in the sample characteristics for the target object that the application refers to, target object feature may include user's
Organ characteristic specifically can also be the supercilium feature that for example may include user, eye feature, lip feature, forehead feature, ear
Portion's feature etc..Further, the application can use neural network model, detect come face organ's feature to user
And it analyzes.It specifically can also be that for example may include use again or, target object feature also may include the limbs feature of user
The finger characteristic at family, palm feature, arm feature, the back of the hand feature etc..Likewise, the application also can use neural network mould
Type is detected and is analyzed come the limbs feature to user.
Further, the application can be by neural network image semantic segmentation model, to identify in video to be processed
At least one target object included by image.Still further, neural network image semantic segmentation model can also be to figure
Each feature in target object as in is classified, and the feature for belonging to same classification is divided into same type, in this way,
Obtained target object feature can be to be made of multiple and different features after image carries out semantic segmentation.
Optionally, for used neural network image semantic segmentation model, in a kind of embodiment, sample can be passed through
This image is trained neural network image semantic segmentation model.Specifically, available sample image, and utilize sample graph
As being trained to preset neural network image semantic segmentation model, the neural network image for obtaining meeting preset condition is semantic
Parted pattern.
Wherein, sample image includes at least one target object feature, and target object feature can be with the embodiment of the present application
In target object feature it is identical.For example, the sample object object features in sample image may include user face other,
Hand limbs etc..
When neural network image semantic segmentation model carries out semantic segmentation processing to sample image, to the picture in sample image
Vegetarian refreshments classification is more accurate, then identifies that the accuracy rate of the tagged object in sample image is higher.Wherein it should be noted that default item
Part can be customized setting.
For example, preset condition can be set are as follows: 70% or more is reached to the classification accuracy of pixel, then, sample graph
As carrying out repetition training to neural network image semantic segmentation model, in neural network image semantic segmentation model to pixel
When classification accuracy reaches 70% or more, then the neural network image semantic segmentation model can be applied in present invention implementation at this time
In example, semantic segmentation processing is carried out to image.
It is available that there is identical aspect ratio with target object type after obtaining target object type in the application
The identification window of example and size, and image is calculated using identification window.Pass through the technical side of application the application
Case can choose matching identification window automatically and carry out depth to image according to the type and area of target object
It practises and calculating.And then can to avoid image recognition it is incorrect caused by influence depth study calculated result the drawbacks of.
It is further alternative, in one of embodiment of the application, (it is based on identification window, to image in S103
Carry out the identification of target object) in, it further include a kind of specific embodiment, as shown in Figure 5, comprising:
S301 obtains target object type.
S302 is based on target object type, obtains identification window.
S303 identifies the target object in image.
S304 intercepts image based on identification window, obtains at least one subgraph, at least one subgraph into
The identification of row target object.
In the application, after getting target identification window, can also further it identify in the picture, target object
Region and target object horizontal direction in the picture.So that it is subsequent according to the area to be tested and horizontal direction,
Using mark identification window, image is identified.
Identification window is pressed horizontal direction and/or vertical direction sliding in the picture, successively intercepts, obtain at least by S305
One subgraph；And/or by horizontal direction and/or vertical direction, multiple identification windows are arranged simultaneously in the picture, respectively
Interception obtains at least one subgraph.
In a kind of possible embodiment of the application, it can also be accomplished by the following way identification window in image
In slided by horizontal direction and/or vertical direction, successively intercept, obtain at least one subgraph；And/or by horizontal direction and/
Or vertical direction, multiple identification windows are arranged simultaneously in the picture, interception obtains at least one subgraph respectively:
Obtain multiple identification windows；
According to multiple identification windows, image is identified, multiple identification windows be apart from area to be tested it is default away from
From, and it is located at the horizontal direction of target object and/or the identification window of vertical direction, the position of multiple identification windows in the picture
It is different.
Optionally, although in order to avoid there is suitable identification window, since identification window does not have automatic coverage goal
The problem of recognition result inaccuracy caused by all profiles of object.In the application, target object can be detected first and is being schemed
Region as in, and multiple horizontal direction target identification windows identical as target object are utilized simultaneously, target object is marked
Note.Wherein it should be noted that the position of each identification window in the picture is different.
By target object be image in teacher face organ for, due to be usually present in the related technology identification window without
The problem of method accurately marks target object completely.Such as the calibration position of identification window does not have it can be seen from Fig. 6 a
The face organ of teacher is labeled completely.To will lead to since the recognition result that image recognition is not in place and occurs is inaccurate
True problem.
In order to solve this problem, multiple target identification windows can be obtained in the application simultaneously, and by multiple identification
Window is repeatedly labeled target object with same level direction around the region of target object.It is illustrated with horizontal direction
Illustrate, as shown in Figure 6 b, it can be seen that target identification window 1, target identification window 2, target identification window 3, target identification window
Mouth 4 is to have multiple identification windows in same level direction in the picture, and each target identification window is apart to be detected
Region pre-determined distance is marked.Further, so as to carry out image recognition to image according to multiple target identification windows subsequent
Result in, successively intercept, obtain at least one subgraph.And/or by horizontal direction and/or vertical direction, by multiple identifications
Window is arranged simultaneously in the picture and is intercepted respectively, to obtain corresponding multiple subgraphs.And then subsequent according to multiple subgraphs
The recognition result of picture selects the most accurate subgraph of recognition result to occur to user.So as to improve the accuracy rate of identification.
In the application, after obtaining target object type, the available target to match with target object type is known
Other window, and area to be tested, horizontal direction to be detected and multiple target identification windows using target object in the picture,
Image is calculated.It, can be according to the area to be detected in the picture of target object by the technical solution of application the application
Domain and horizontal direction to be detected, using multiple target identification windows to image recognition.And then it can be to avoid due to identification window
The drawbacks of reducing recognition accuracy caused by labeling position inaccuracy in the picture.
In another embodiment of the application, as shown in fig. 7, the application also provides a kind of pattern recognition device,
The device includes the first acquisition module 401, and second obtains module 402, computing module 403, wherein
First obtains module 401, is configured as obtaining target object type, the target object type is for being reflected in figure
As in, the classification of target object；
Second obtains module 402, is configured as obtaining target identification window, the mesh based on the target object type
Marking identification window is the identification window to match with the target object type；
Computing module 403 is configured as calculating described image based on the target identification window.
In the application, after obtaining target object type, the available target to match with target object type is known
Other window, and area to be tested, horizontal direction to be detected and multiple target identification windows using target object in the picture,
Image is calculated.It, can be according to the area to be detected in the picture of target object by the technical solution of application the application
Domain and horizontal direction to be detected carry out deep learning calculating to image using multiple target identification windows.And then it can be to avoid
The drawbacks of due to reducing recognition accuracy caused by the labeling position inaccuracy of identification window in the picture.
Optionally, in the another embodiment of the application, it further includes that determination unit obtains that second, which obtains module 402,
Take unit, in which:
Determination unit is configured as determining the Aspect Ratio of the target object based on the target object type；
Acquiring unit is configured to determine that the Aspect Ratio of the identification window.
In the another embodiment of the application, it further includes determination unit that second, which obtains module 402, acquiring unit,
Wherein:
Acquiring unit is configured to determine that shared regional percentage of the target object in described image；
Determination unit is configured as calculating the area of the identification window based on the shared regional percentage；
Acquiring unit is configured as area and the Aspect Ratio based on the identification window, obtains the identification
Window.
In the another embodiment of the application, it further includes determination unit that second, which obtains module 402, in which:
Determination unit is configured as being based on the target when the target object quantity in described image is multiple
The type of the maximum target object of area, determines identification window in object.
In the another embodiment of the application, computing module 403 further includes recognition unit, computing unit, in which:
Recognition unit is configured as intercepting described image based on the identification window, obtains at least one subgraph
Picture carries out the identification of the target object at least one described subgraph.
In the another embodiment of the application, computing module 403 further includes recognition unit, computing unit, in which:
Acquiring unit is configured as the identification window is sliding by horizontal direction and/or vertical direction in described image
It is dynamic, it successively intercepts, obtains at least one described subgraph；And/or
By horizontal direction and/or vertical direction, multiple identification windows are arranged simultaneously in described image, are cut respectively
Obtain at least one described subgraph.
In the another embodiment of the application, computing module 403, further includes:
Computing module is configured as knowing based on trained neural network image semantic segmentation model in advance and the target
Other window, calculates described image.
Fig. 8 is the logical construction block diagram of a kind of electronic equipment shown according to an exemplary embodiment.For example, electronic equipment
500 can be mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, and medical treatment is set
It is standby, body-building equipment, personal digital assistant etc..
Referring to Fig. 8, electronic equipment 500 may include following one or more components: processor 501 and memory 502.
Processor 501 may include one or more processing cores, such as 4 core processors, 8 core processors etc..Place
Reason device 501 can use DSP (Digital Signal Processing, Digital Signal Processing), FPGA (Field-
Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array, may be programmed
Logic array) at least one of example, in hardware realize.Processor 501 also may include primary processor and coprocessor, master
Processor is the processor for being handled data in the awake state, also referred to as CPU (Central Processing
Unit, central processing unit)；Coprocessor is the low power processor for being handled data in the standby state.?
In some embodiments, processor 501 can be integrated with GPU (Graphics Processing Unit, image processor),
GPU is used to be responsible for the rendering and drafting of content to be shown needed for display screen.In some embodiments, processor 501 can also be wrapped
AI (Artificial Intelligence, artificial intelligence) processor is included, the AI processor is for handling related machine learning
Calculating operation.
Memory 502 may include one or more computer readable storage mediums, which can
To be non-transient.Memory 502 may also include high-speed random access memory and nonvolatile memory, such as one
Or multiple disk storage equipments, flash memory device.In some embodiments, the non-transient computer in memory 502 can
Storage medium is read for storing at least one instruction, at least one instruction performed by processor 501 for realizing this Shen
Please in embodiment of the method provide interaction special efficacy scaling method.
In some embodiments, electronic equipment 500 is also optional includes: peripheral device interface 503 and at least one periphery
Equipment.It can be connected by bus or signal wire between processor 501, memory 502 and peripheral device interface 503.It is each outer
Peripheral equipment can be connected by bus, signal wire or circuit board with peripheral device interface 503.Specifically, peripheral equipment includes: to penetrate
At least one in frequency circuit 504, touch display screen 505, camera 506, voicefrequency circuit 507, positioning component 508 and power supply 509
Kind.
Peripheral device interface 503 can be used for I/O (Input/Output, input/output) is relevant outside at least one
Peripheral equipment is connected to processor 501 and memory 502.In some embodiments, processor 501, memory 502 and peripheral equipment
Interface 503 is integrated on same chip or circuit board；In some other embodiments, processor 501, memory 502 and outer
Any one or two in peripheral equipment interface 503 can realize on individual chip or circuit board, the present embodiment to this not
It is limited.
Radio circuit 504 is for receiving and emitting RF (Radio Frequency, radio frequency) signal, also referred to as electromagnetic signal.It penetrates
Frequency circuit 504 is communicated by electromagnetic signal with communication network and other communication equipments.Radio circuit 504 turns electric signal
It is changed to electromagnetic signal to be sent, alternatively, the electromagnetic signal received is converted to electric signal.Optionally, radio circuit 504 wraps
It includes: antenna system, RF transceiver, one or more amplifiers, tuner, oscillator, digital signal processor, codec chip
Group, user identity module card etc..Radio circuit 504 can be carried out by least one wireless communication protocol with other terminals
Communication.The wireless communication protocol includes but is not limited to: Metropolitan Area Network (MAN), each third generation mobile communication network (2G, 3G, 4G and 5G), wireless office
Domain net and/or WiFi (Wireless Fidelity, Wireless Fidelity) network.In some embodiments, radio circuit 504 may be used also
To include the related circuit of NFC (Near Field Communication, wireless near field communication), the application is not subject to this
It limits.
Display screen 505 is for showing UI (User Interface, user interface).The UI may include figure, text, figure
Mark, video and its their any combination.When display screen 505 is touch display screen, display screen 505 also there is acquisition to show
The ability of the touch signal on the surface or surface of screen 505.The touch signal can be used as control signal and be input to processor
501 are handled.At this point, display screen 505 can be also used for providing virtual push button and/or dummy keyboard, also referred to as soft button and/or
Soft keyboard.In some embodiments, display screen 505 can be one, and the front panel of electronic equipment 500 is arranged；In other realities
It applies in example, display screen 505 can be at least two, be separately positioned on the different surfaces of electronic equipment 500 or in foldover design；?
In still other embodiments, display screen 505 can be flexible display screen, is arranged on the curved surface of electronic equipment 500 or folds
On face.Even, display screen 505 can also be arranged to non-rectangle irregular figure, namely abnormity screen.Display screen 505 can be adopted
With LCD (Liquid Crystal Display, liquid crystal display), (Organic Light-Emitting Diode, has OLED
Machine light emitting diode) etc. materials preparation.
CCD camera assembly 506 is for acquiring image or video.Optionally, CCD camera assembly 506 include front camera and
Rear camera.In general, the front panel of terminal is arranged in front camera, the back side of terminal is arranged in rear camera.One
In a little embodiments, rear camera at least two is main camera, depth of field camera, wide-angle camera, focal length camera shooting respectively
Any one in head, to realize that main camera and the fusion of depth of field camera realize background blurring function, main camera and wide-angle
Camera fusion realizes that pan-shot and VR (Virtual Reality, virtual reality) shooting function or other fusions are clapped
Camera shooting function.In some embodiments, CCD camera assembly 506 can also include flash lamp.Flash lamp can be monochromatic warm flash lamp,
It is also possible to double-colored temperature flash lamp.Double-colored temperature flash lamp refers to the combination of warm light flash lamp and cold light flash lamp, can be used for not
With the light compensation under colour temperature.
Voicefrequency circuit 507 may include microphone and loudspeaker.Microphone is used to acquire the sound wave of user and environment, and will
Sound wave, which is converted to electric signal and is input to processor 501, to be handled, or is input to radio circuit 504 to realize voice communication.
For stereo acquisition or the purpose of noise reduction, microphone can be separately positioned on the different parts of electronic equipment 500 to be multiple.
Microphone can also be array microphone or omnidirectional's acquisition type microphone.Loudspeaker is then used to that processor 501 or radio frequency will to be come from
The electric signal of circuit 504 is converted to sound wave.Loudspeaker can be traditional wafer speaker, be also possible to piezoelectric ceramics loudspeaking
Device.When loudspeaker is piezoelectric ceramic loudspeaker, the audible sound wave of the mankind can be not only converted electrical signals to, can also be incited somebody to action
Electric signal is converted to the sound wave that the mankind do not hear to carry out the purposes such as ranging.In some embodiments, voicefrequency circuit 507 can be with
Including earphone jack.
Positioning component 508 is used for the current geographic position of Positioning Electronic Devices 500, to realize navigation or LBS (Location
Based Service, location based service).Positioning component 508 can be the GPS (Global based on the U.S.
Positioning System, global positioning system), the dipper system of China, Russia Gray receive this system or European Union
The positioning component of Galileo system.
Power supply 509 is used to be powered for the various components in electronic equipment 500.Power supply 509 can be alternating current, direct current
Electricity, disposable battery or rechargeable battery.When power supply 509 includes rechargeable battery, which can support wired
Charging or wireless charging.The rechargeable battery can be also used for supporting fast charge technology.
In some embodiments, electronic equipment 500 further includes having one or more sensors 510.The one or more passes
Sensor 510 includes but is not limited to: acceleration transducer 511, gyro sensor 512, pressure sensor 513, fingerprint sensor
514, optical sensor 515 and proximity sensor 516.
Acceleration transducer 511 can detecte the acceleration in three reference axis of the coordinate system established with electronic equipment 500
Spend size.For example, acceleration transducer 511 can be used for detecting component of the acceleration of gravity in three reference axis.Processor
The 501 acceleration of gravity signals that can be acquired according to acceleration transducer 511, control touch display screen 505 with transverse views or
Longitudinal view carries out the display of user interface.Acceleration transducer 511 can be also used for game or the exercise data of user
Acquisition.
Gyro sensor 512 can detecte body direction and the rotational angle of electronic equipment 500, gyro sensor
512 can cooperate with acquisition user to act the 3D of electronic equipment 500 with acceleration transducer 511.Processor 501 is according to gyroscope
The data that sensor 512 acquires, may be implemented following function: action induction (for example changed according to the tilt operation of user
UI), image stabilization, game control and inertial navigation when shooting.
The lower layer of side frame and/or touch display screen 505 in electronic equipment 500 can be set in pressure sensor 513.When
When the side frame of electronic equipment 500 is arranged in pressure sensor 513, user can detecte to the gripping signal of electronic equipment 500,
Right-hand man's identification or prompt operation are carried out according to the gripping signal that pressure sensor 513 acquires by processor 501.Work as pressure sensing
When the lower layer of touch display screen 505 is arranged in device 513, grasped by processor 501 according to pressure of the user to touch display screen 505
Make, realization controls the operability control on the interface UI.Operability control include button control, scroll bar control,
At least one of icon control, menu control.
Fingerprint sensor 514 is used to acquire the fingerprint of user, collected according to fingerprint sensor 514 by processor 501
The identity of fingerprint recognition user, alternatively, by fingerprint sensor 514 according to the identity of collected fingerprint recognition user.It is identifying
When the identity of user is trusted identity out, the user is authorized to execute relevant sensitive operation, the sensitive operation packet by processor 501
Include solution lock screen, check encryption information, downloading software, payment and change setting etc..Electronics can be set in fingerprint sensor 514
Front, the back side or the side of equipment 500.When being provided with physical button or manufacturer Logo on electronic equipment 500, fingerprint sensor
514 can integrate with physical button or manufacturer Logo.
Optical sensor 515 is for acquiring ambient light intensity.In one embodiment, processor 501 can be according to optics
The ambient light intensity that sensor 515 acquires controls the display brightness of touch display screen 505.Specifically, when ambient light intensity is higher
When, the display brightness of touch display screen 505 is turned up；When ambient light intensity is lower, the display for turning down touch display screen 505 is bright
Degree.In another embodiment, the ambient light intensity that processor 501 can also be acquired according to optical sensor 515, dynamic adjust
The acquisition parameters of CCD camera assembly 506.
Proximity sensor 516, also referred to as range sensor are generally arranged at the front panel of electronic equipment 500.Proximity sensor
516 for acquiring the distance between the front of user Yu electronic equipment 500.In one embodiment, when proximity sensor 516 is examined
When measuring the distance between the front of user and electronic equipment 500 and gradually becoming smaller, touch display screen 505 is controlled by processor 501
Breath screen state is switched to from bright screen state；When proximity sensor 516 detect between user and the front of electronic equipment 500 away from
When from becoming larger, touch display screen 505 being controlled by processor 501 and is switched to bright screen state from breath screen state.
It will be understood by those skilled in the art that structure shown in Fig. 8 does not constitute the restriction to electronic equipment 500, it can
To include perhaps combining certain components than illustrating more or fewer components or being arranged using different components.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instruction, example are additionally provided
It such as include the memory 504 of instruction, above-metioned instruction can be executed by the processor 520 of electronic equipment 500 to complete above-mentioned image and know
Other method, this method comprises: determining target object type, the target object type is for reflecting in the picture, the target
The classification of object；Based on the target object type, identification window, the identification window and the target object type phase are determined
Matching；Based on the identification window, the identification of the target object is carried out to described image.Optionally, above-metioned instruction can be with
It is executed as the processor 520 of electronic equipment 500 to complete other steps involved in the above exemplary embodiments.For example, institute
State non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk and
Optical data storage devices etc..
In the exemplary embodiment, a kind of application program/computer program product is additionally provided, including one or more refers to
It enables, which can be executed by the processor 520 of electronic equipment 500, should to complete above-mentioned image-recognizing method
Method comprises determining that target object type, and the target object type is for reflecting in the picture, the class of the target object
Not；Based on the target object type, determine that identification window, the identification window match with the target object type；Base
In the identification window, the identification of the target object is carried out to described image.Optionally, above-metioned instruction can also be set by electronics
Standby 500 processor 520 is executed to complete other steps involved in the above exemplary embodiments.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the application
Its embodiment.This application is intended to cover any variations, uses, or adaptations of the application, these modifications, purposes or
Person's adaptive change follows the general principle of the application and including the undocumented common knowledge in the art of the application
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the application are by following
Claim is pointed out.
It should be understood that the application is not limited to the precise structure that has been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.Scope of the present application is only limited by the accompanying claims.
Claims (10)
1. a kind of image-recognizing method characterized by comprising
Determine target object type, the target object type is for reflecting in the picture, the classification of the target object；
Based on the target object type, determine that identification window, the identification window match with the target object type；
Based on the identification window, the identification of the target object is carried out to described image.
2. the method as described in claim 1, which is characterized in that it is described to be based on the target object type, determine identification window,
Include:
Based on the target object type, the Aspect Ratio of the target object is determined；
Based on the Aspect Ratio, the Aspect Ratio of the identification window is determined.
3. method according to claim 2, which is characterized in that it is described to be based on the Aspect Ratio, determine the length of identification window
Wide ratio, comprising:
Determine shared regional percentage of the target object in described image；
Based on the shared regional percentage, the area of the identification window is calculated；
Area and the Aspect Ratio based on the identification window, obtain the identification window.
4. the method as described in claim 1, which is characterized in that when the target object quantity in described image is multiple
When, based on the type of the maximum target object of area in the target object, determine identification window.
5. the method as described in claim 1, which is characterized in that it is described to be based on the identification window, described image is known
Other places reason, comprising:
Described image is intercepted based on the identification window, obtains at least one subgraph, at least one described subgraph
Identification as carrying out the target object.
6. method as claimed in claim 5, which is characterized in that described to be cut based on the identification window to described image
It takes, obtains at least one subgraph, comprising:
The identification window is slided in described image by horizontal direction and/or vertical direction, is successively intercepted, obtain it is described extremely
A few subgraph；And/or
By horizontal direction and/or vertical direction, multiple identification windows are arranged simultaneously in described image, are intercepted respectively
To at least one described subgraph.
7. the method as described in claim 1, which is characterized in that it is described to be based on the identification window, institute is carried out to described image
State the identification of target object, comprising:
Based on trained neural network image semantic segmentation model and the identification window in advance, described in described image
Target object is identified.
8. a kind of pattern recognition device characterized by comprising
Module is obtained, is configured to determine that target object type, the target object type is for reflecting in the picture, the mesh
Mark the classification of object；
Determining module is configured as determining identification window, the identification window and the target based on the target object type
Object type matches；
Identification module is configured as carrying out described image the identification of the target object based on the identification window.
9. a kind of electronic equipment characterized by comprising
Memory, for storing executable instruction；And
It is any in claim 1-7 to complete to execute the executable instruction for showing with the memory for display
The operation of described image recognition methods.
10. a kind of computer readable storage medium, for storing computer-readable instruction, which is characterized in that described instruction
It is performed the operation that perform claim requires any described image recognition methods in 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910475041.9A CN110334736A (en) | 2019-06-03 | 2019-06-03 | Image-recognizing method, device, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910475041.9A CN110334736A (en) | 2019-06-03 | 2019-06-03 | Image-recognizing method, device, electronic equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110334736A true CN110334736A (en) | 2019-10-15 |
Family
ID=68140634
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910475041.9A Pending CN110334736A (en) | 2019-06-03 | 2019-06-03 | Image-recognizing method, device, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110334736A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110927167A (en) * | 2019-10-31 | 2020-03-27 | 北京海益同展信息科技有限公司 | Egg detection method and device, electronic equipment and storage medium |
| CN111931725A (en) * | 2020-09-23 | 2020-11-13 | 北京无垠创新科技有限责任公司 | Human body action recognition method, device and storage medium |
| CN112733985A (en) * | 2021-01-27 | 2021-04-30 | 昆山善思光电科技有限公司 | Method for automatically identifying large and small material trays |
| CN113221604A (en) * | 2020-02-05 | 2021-08-06 | 北京京东乾石科技有限公司 | Target identification method and device, storage medium and electronic equipment |
| CN114526722A (en) * | 2021-12-31 | 2022-05-24 | 易图通科技（北京）有限公司 | Map alignment processing method and device and readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389813A (en) * | 2015-10-30 | 2016-03-09 | 上海联影医疗科技有限公司 | Medical image organ recognition method and segmentation method |
| CN105654461A (en) * | 2014-12-02 | 2016-06-08 | 中国科学院沈阳自动化研究所 | Machine vision detection method for fracture of multi-bundle conductor spacer |
| CN109214336A (en) * | 2018-09-05 | 2019-01-15 | 高新兴科技集团股份有限公司 | A kind of vehicle window marker detection method and device |
| CN109254775A (en) * | 2018-08-30 | 2019-01-22 | 广州酷狗计算机科技有限公司 | Image processing method, terminal and storage medium based on face |
| CN109657537A (en) * | 2018-11-05 | 2019-04-19 | 北京达佳互联信息技术有限公司 | Image-recognizing method, system and electronic equipment based on target detection |
-
2019
- 2019-06-03 CN CN201910475041.9A patent/CN110334736A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105654461A (en) * | 2014-12-02 | 2016-06-08 | 中国科学院沈阳自动化研究所 | Machine vision detection method for fracture of multi-bundle conductor spacer |
| CN105389813A (en) * | 2015-10-30 | 2016-03-09 | 上海联影医疗科技有限公司 | Medical image organ recognition method and segmentation method |
| CN109254775A (en) * | 2018-08-30 | 2019-01-22 | 广州酷狗计算机科技有限公司 | Image processing method, terminal and storage medium based on face |
| CN109214336A (en) * | 2018-09-05 | 2019-01-15 | 高新兴科技集团股份有限公司 | A kind of vehicle window marker detection method and device |
| CN109657537A (en) * | 2018-11-05 | 2019-04-19 | 北京达佳互联信息技术有限公司 | Image-recognizing method, system and electronic equipment based on target detection |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110927167A (en) * | 2019-10-31 | 2020-03-27 | 北京海益同展信息科技有限公司 | Egg detection method and device, electronic equipment and storage medium |
| CN113221604A (en) * | 2020-02-05 | 2021-08-06 | 北京京东乾石科技有限公司 | Target identification method and device, storage medium and electronic equipment |
| CN111931725A (en) * | 2020-09-23 | 2020-11-13 | 北京无垠创新科技有限责任公司 | Human body action recognition method, device and storage medium |
| CN111931725B (en) * | 2020-09-23 | 2023-10-13 | 北京无垠创新科技有限责任公司 | Human motion recognition method, device and storage medium |
| CN112733985A (en) * | 2021-01-27 | 2021-04-30 | 昆山善思光电科技有限公司 | Method for automatically identifying large and small material trays |
| CN114526722A (en) * | 2021-12-31 | 2022-05-24 | 易图通科技（北京）有限公司 | Map alignment processing method and device and readable storage medium |
| CN114526722B (en) * | 2021-12-31 | 2024-05-24 | 易图通科技（北京）有限公司 | Map alignment processing method and device and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110334736A (en) | Image-recognizing method, device, electronic equipment and medium | |
| CN109829456A (en) | Image-recognizing method, device and terminal | |
| CN108710525A (en) | Map methods of exhibiting, device, equipment and storage medium in virtual scene | |
| CN110046018A (en) | Show method, apparatus, electronic equipment and the medium of message | |
| CN110113659A (en) | Generate method, apparatus, electronic equipment and the medium of video | |
| CN113763228B (en) | Image processing method, device, electronic equipment and storage medium | |
| CN109886208B (en) | Object detection method and device, computer equipment and storage medium | |
| CN108363982B (en) | Method and device for determining number of objects | |
| CN110796005A (en) | Method, device, electronic equipment and medium for online teaching monitoring | |
| CN110532170A (en) | Build method, apparatus, electronic equipment and the medium of test environment | |
| CN110348318A (en) | Image-recognizing method, device, electronic equipment and medium | |
| CN110535820A (en) | For the classification method of malice domain name, device, electronic equipment and medium | |
| CN110675473B (en) | Method, device, electronic equipment and medium for generating GIF dynamic diagram | |
| CN108852286A (en) | Show the method, apparatus and terminal of backbone measurement data | |
| CN110288689A (en) | The method and apparatus that electronic map is rendered | |
| CN109992685A (en) | A kind of method and device of retrieving image | |
| CN111932604A (en) | Method and device for measuring human ear characteristic distance | |
| CN109754439A (en) | Scaling method, device, electronic equipment and medium | |
| CN110378318A (en) | Character recognition method, device, computer equipment and storage medium | |
| CN112860046B (en) | Method, device, electronic equipment and medium for selecting operation mode | |
| CN110853124B (en) | Method, device, electronic equipment and medium for generating GIF dynamic diagram | |
| CN110110142A (en) | Method for processing video frequency, device, electronic equipment and medium | |
| CN110795660B (en) | Data analysis method, data analysis device, electronic device, and medium | |
| CN112925922A (en) | Method, device, electronic equipment and medium for obtaining address | |
| CN110109770A (en) | Adjustment method, device, electronic equipment and 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191015 |