CN112966606B - Image recognition method, related device and computer program product - Google Patents
Image recognition method, related device and computer program product Download PDFInfo
- Publication number
- CN112966606B CN112966606B CN202110245509.2A CN202110245509A CN112966606B CN 112966606 B CN112966606 B CN 112966606B CN 202110245509 A CN202110245509 A CN 202110245509A CN 112966606 B CN112966606 B CN 112966606B
- Authority
- CN
- China
- Prior art keywords
- image
- clear
- identified
- same
- reference image
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000004590 computer program Methods 0.000 title abstract description 15
- 230000004044 response Effects 0.000 claims abstract description 19
- 238000004891 communication Methods 0.000 claims description 59
- 238000010586 diagram Methods 0.000 claims description 22
- 238000012545 processing Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/457—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by analysing connectivity, e.g. edge linking, connected component analysis or slices
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the application discloses an image recognition method, an image recognition device, electronic equipment, a computer readable storage medium and a computer program product, relates to the technical field of computer vision, and can be applied to intelligent traffic scenes. One embodiment of the method comprises the following steps: and determining a non-clear object and a clear object with definition larger than that of the non-clear object in the image to be identified, acquiring a reference image containing the same-class object and the clear object of the non-clear object, and determining that the same-class object and the non-clear object are the same object in response to the same position relationship of a first object group consisting of the non-clear object and the clear object in the image to be identified and the position relationship of a second object group consisting of the same-class object and the clear object in the reference image. According to the method, the reference picture is found through the clear objects in the image to be identified, so that the non-clear objects in the image to be identified are identified in an auxiliary mode through high-definition content contained in the reference image.
Description
Technical Field
The present invention relates to the field of image processing technology, and in particular, to the field of computer vision technology, which is applicable to intelligent traffic scenes, and particularly, to an image recognition method, an image recognition device, an electronic device, a computer readable storage medium, and a computer program product.
Background
In recent years, in order to acquire scene information and monitor certain scenes in real time, the use frequency of cameras is gradually increased, and particularly in projects such as smart city construction, a large amount of cameras are required to acquire image data for scene feedback in order to achieve the purposes of large data acquisition and smart scheduling.
In the prior art, in order to improve the processing efficiency of image data, the image data is clustered according to whether the image data contains specific content.
Disclosure of Invention
The embodiment of the application provides an image identification method, an image identification device, electronic equipment, a computer readable storage medium and a computer program product.
In a first aspect, an embodiment of the present application provides an image recognition method, including: determining an unclear object and a clear object in an image to be identified; wherein the definition of the non-clear object is less than the definition of the clear object; acquiring a reference image of the same-class object containing the non-clear object and the clear object; wherein the definition of the same class object is greater than the definition of the non-clear object; determining that the same-class object and the non-clear object are the same object in response to the fact that the position relation of the first object group in the image to be identified is the same as the position relation of the second object group in the reference image; wherein the first object group is composed of the non-distinct object and the distinct object, and the second object group is composed of the same class object and the distinct object.
In a second aspect, an embodiment of the present application proposes an image recognition apparatus, including: an object determining unit configured to determine an unclear object and a clear object in an image to be recognized; wherein the definition of the non-clear object is less than the definition of the clear object; a reference image acquisition unit configured to acquire a reference image of a same-class object including the non-sharp object and the sharp object; wherein the definition of the same class object is greater than the definition of the non-clear object; a content discriminating unit configured to determine that the same-category object and the non-clear object are the same object in response to a positional relationship of the first object group in the image to be recognized being the same as a positional relationship of the second object group in the reference image; wherein the first object group is composed of the non-distinct object and the distinct object, and the second object group is composed of the same class object and the distinct object.
In a third aspect, an embodiment of the present application provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to implement the image recognition method as described in any one of the implementations of the first aspect when executed.
In a fourth aspect, embodiments of the present application provide a non-transitory computer-readable storage medium storing computer instructions for enabling a computer to implement an image recognition method as described in any one of the implementations of the first aspect when executed.
In a fifth aspect, embodiments of the present application provide a computer program product comprising a computer program which, when executed by a processor, is capable of implementing an image recognition method as described in any of the implementations of the first aspect.
The image recognition method, the device, the electronic equipment, the computer readable storage medium and the computer program product provided by the embodiment of the application acquire a reference image containing a same-class object and a clear object of the non-clear object after determining that the non-clear object and the clear object with the definition larger than the non-clear object in an image to be recognized, and determine that the same-class object and the non-clear object are the same object in response to the fact that the position relationship of a first object group consisting of the non-clear object and the clear object in the image to be recognized is the same as the position relationship of a second object group consisting of the same-class object and the clear object in the reference image.
The method and the device determine the reference image based on the clear objects of the non-clear objects contained in the image to be identified, so that the non-clear objects in the image to be identified can be identified in an auxiliary mode through the high-definition content contained in the reference image, the problem that the non-clear objects in the image to be identified cannot be identified due to the fact that the non-clear objects in the image to be identified are blocked, the angles are not correct, the shooting is not clear and the like is solved, and therefore the identification accuracy and the identification efficiency of image identification are improved.
It should be understood that the description of this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the following drawings, in which:
FIG. 1 is an exemplary system architecture in which the present application may be applied;
fig. 2 is a flowchart of an image recognition method according to an embodiment of the present application;
FIG. 3 is a flowchart of another image recognition method according to an embodiment of the present application;
FIGS. 4-1 and 4-2 are schematic diagrams of an image to be identified and a reference image under an application scenario according to an embodiment of the present application;
Fig. 5 is a block diagram of an image recognition device according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device suitable for executing an image recognition method according to an embodiment of the present application.
Detailed Description
Exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness. It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other.
In addition, in the technical scheme disclosed in the application, the acquisition (for example, the image including the face object related later in the application), storage, application and the like of the related personal information of the user all conform to the regulations of related laws and regulations and do not violate the popular regulations.
FIG. 1 illustrates an exemplary system architecture 100 to which embodiments of the image recognition methods, apparatus, electronic devices, and computer-readable storage media of the present application may be applied.
As shown in fig. 1, a system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.
The user can interact with the server 105 via the network 104 using the terminal devices 101, 102, 103 to receive or transmit images or the like. Various applications for implementing information communication between the terminal devices 101, 102, 103 and the server 105, such as a remote image clustering application, an image recognition application, an instant messaging application, and the like, may be installed on the terminal devices.
The terminal devices 101, 102, 103 and the server 105 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices with display screens, including but not limited to smartphones, tablets, laptop and desktop computers, etc.; when the terminal devices 101, 102, 103 are software, they may be installed in the above-listed electronic devices, which may be implemented as a plurality of software or software modules, or may be implemented as a single software or software module, which is not particularly limited herein. When the server 105 is hardware, it may be implemented as a distributed server cluster formed by a plurality of servers, or may be implemented as a single server; when the server is software, the server may be implemented as a plurality of software or software modules, or may be implemented as a single software or software module, which is not particularly limited herein.
The server 105 can provide various services through various built-in applications, and for example, an image recognition class that can provide image content recognition for a user, the server 105 can achieve the following effects when running the image recognition class application: firstly, determining a non-clear object and a clear object from an image to be identified containing the non-clear object in terminal equipment 101, 102 and 103 through a network 104, determining the non-clear object and the clear object from the image to be identified, wherein the definition of the clear object is larger than that of the non-clear object, acquiring a reference image containing a same-class object and the clear object of the non-clear object, wherein the definition of the same-class object is larger than that of the non-clear object, and determining the same-class object and the non-clear object as the same object in response to the fact that the position relation of a first object group consisting of the non-clear object and the clear object in the image to be identified is the same as the position relation of a second object group consisting of the same-class object and the clear object in the reference image, and further transmitting the candidate image to the terminal equipment 101, 102 and 103 to provide the reference image which also contains the non-clear object with high definition for assisting the user in identification.
It is to be noted that the image to be recognized may be stored in advance in the server 105 in various ways, in addition to being acquired from the terminal apparatuses 101, 102, 103 through the network 104. Thus, when the server 105 detects that such data has been stored locally (e.g., a number of images that remain prior to beginning processing, or images to be identified that were extracted from a local database for purposes of image clustering), it may choose to retrieve such data directly from the local, in which case the exemplary system architecture 100 may also exclude the terminal devices 101, 102, 103 and network 104.
Since image recognition requires more computing resources and stronger computing power, the image recognition method provided in the subsequent embodiments of the present application is generally executed by the server 105 having stronger computing power and more computing resources, and accordingly, the image recognition device is also generally disposed in the server 105. However, it should be noted that, when the terminal devices 101, 102, 103 also have the required computing capability and computing resources, the terminal devices 101, 102, 103 may also complete each operation performed by the server 105 through the image recognition application installed thereon, and further output the same result as the server 105. Particularly, in the case where a plurality of terminal devices having different computing capabilities exist at the same time, when the image recognition application determines that the terminal device has a higher computing capability and more computing resources remain, the terminal device may perform the above-mentioned computation, so that the computing pressure of the server 105 is appropriately reduced, and accordingly, the image recognition device may be provided in the terminal devices 101, 102, 103. In this case, the exemplary system architecture 100 may also not include the server 105 and the network 104.
It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
Referring to fig. 2, fig. 2 is a flowchart of an image recognition method according to an embodiment of the present application, where the flowchart 200 includes the following steps:
in step 201, a non-sharp object and a sharp object in an image to be identified are determined.
In the present embodiment, an image to be recognized including a non-sharp object is acquired by an execution subject of an image recognition method (for example, a server 105 shown in fig. 1), and the non-sharp object and the sharp object are respectively determined from the acquired image to be recognized.
The clear object is an object with definition greater than that of the non-clear object, the clear object can be directly determined as an environmental object in an image to be identified, such as an object of a street lamp, a billboard, a bench and the like, or the corresponding clear object can be determined according to the relevance of the non-clear object and the non-clear object according to the difference of the non-clear object, for example, when the non-clear object is a driver, the clear object can be correspondingly selected as license plate information, vehicle appearance information and the like.
It should be understood that in order to more accurately determine scene information and restore scene conditions contained in an image to be identified in practice, when a distinct object corresponding to a non-distinct object is selected, a distinct object with high relevance and high relevance to the non-distinct object is generally selected so as to facilitate improvement of the reference value of the distinct object.
Further, when determining the non-clear object contained in the image to be identified, a corresponding threshold may be set in advance according to the requirement of the image identification precision, when the object in the image to be identified cannot meet the definition requirement corresponding to the threshold, the object is determined to be the non-clear object, and similarly, the corresponding threshold may also be determined according to the type of the object contained in the image to be identified, and when the object in the image to be identified cannot meet the definition requirement corresponding to the threshold, the object is determined to be the non-clear object.
In practice, when a plurality of different clear objects exist, corresponding definition thresholds can be set according to the type information of each clear object, so that clear objects with preferable quality can be screened out conveniently, and a better image recognition effect is achieved. For example, if the clear object is license plate information, the corresponding definition threshold requirement needs to be satisfied to smoothly extract the Chinese characters, letters and digital contents in the license plate information, and if the clear object is billboard information, the corresponding definition threshold requirement needs to be satisfied to extract certain characteristic information in the billboard.
It should be noted that the image to be identified may be obtained directly from the local storage device by the execution body, or may be obtained directly from an image acquisition device connected to the execution body, or may be obtained from a non-local storage device (for example, the terminal devices 101, 102, 103 shown in fig. 1). The local storage device may be a data storage module, such as a server hard disk, provided in the execution body, in which case the image to be identified may be read quickly locally; the non-local storage device may also be any other electronic device arranged to store data, such as some user terminals or the like, in which case the executing entity may acquire the desired image to be identified by sending an acquisition command to the electronic device.
Step 202, obtaining a reference image of the same class object and the clear object containing the non-clear object.
In this embodiment, after determining the non-clear object and the clear object based on the image to be identified in the above step 201, image recall is performed based on the clear object, and feature information of the non-clear object such as appearance information and attribute information of the non-clear object is parsed and acquired, so that an image of the same class object including the non-clear object is determined as a reference image from the recalled image.
It should be understood that the sharpness of the same-class image of the non-sharp object contained in the finally determined reference image should be greater than the sharpness of the non-sharp image in the image to be identified, so that when the same-class object and the non-sharp object are determined to be the same object, the non-sharp object is identified in an auxiliary manner by the content of the same-class object.
For example, after the non-clear object is resolved, it is determined that the non-clear object is an "X-card blue car", an image including the "X-card blue car" may be determined as a reference image, in practice, the similarity of the same class object corresponding to the non-clear object may be set correspondingly according to the requirement, for example, in this example, in order to obtain more reference images, the non-clear object may be identified as a "blue car" correspondingly.
In some optional embodiments of the present application, recall of the reference image may also be performed by using the clear object and the non-clear object in combination after determining the clear object and the non-clear object, so as to improve the reference value of the recalled image.
In step 203, the same object is determined as the same object in the same category of objects and the non-distinct object in response to the same positional relationship of the first object group in the image to be identified as the second object group in the reference image.
In this embodiment, after determining the reference image, the positional relationship of the first object group formed by the clear object and the non-clear object in the image to be identified and the positional relationship of the second object group formed by the clear object and the same category object in the reference image are respectively determined, so as to determine whether the image to be identified and the reference image are obtained for the same scene content, and under the condition that the positional relationship of the first object group in the image to be identified and the positional relationship of the second object group in the reference image are the same or the approximation degree meets the preset condition, the image to be identified and the reference image can be correspondingly determined to be different images obtained based on the same scene content, so as to determine whether the same category object and the non-clear object are the same object by using the principle of scene restoration.
According to the image recognition method, the reference image is determined based on the clear objects of the non-clear objects contained in the image to be recognized, so that the non-clear objects in the image to be recognized can be recognized in an auxiliary mode through the high-definition content contained in the reference image, the problem that the non-clear objects in the image to be recognized cannot be recognized due to the fact that the non-clear objects in the image to be recognized are blocked, the angles are not correct, the shooting is not clear and the like is solved, and therefore recognition accuracy and recognition efficiency of image recognition are improved.
In some other embodiments of the present application, in order to further combine with the actual needs of the user to perform the identification of the image content, so as to improve the efficiency of the image identification process and improve the use value of the identification result for the user, the specific image identification may be further implemented based on the image to be identified and the corresponding indication information uploaded by the user through the user terminal, where in this case, determining the non-clear object and the clear object in the image to be identified may specifically include:
the method comprises the steps of obtaining an image to be identified and object indication information aiming at the image to be identified, which are uploaded by a user through a user terminal, wherein the indication information is marked with an unclear object in the image to be identified, so that the execution main body can conveniently determine the unclear object according to the object indication information, and determine an object with definition greater than that of the unclear object in the image to be identified as a clear object according to definition determination identification of the unclear object.
In practice, the indication information of the user can also contain the non-clear object and the clear object appointed by the user at the same time, so that the image recognition can be performed directly according to the non-clear object and the clear object appointed by the user, and the problem of poor user experience caused by that the definition of the preset non-clear object and the definition of the clear object can not meet the actual requirement of the user is avoided.
In some other embodiments of the present application, the reference image may be further pushed to the user terminal in response to the reference image including more than a preset number of the distinct objects; and re-determining the reference image according to the clear object specification information returned by the user terminal. When it is determined that the number of clear objects included in the reference image exceeds the preset number, in order to facilitate a user to combine specific requirements and determine a clear object with a higher value, resource waste and slow response caused by image recognition for a clear object with a lower value are reduced, the executing body pushes the reference image to a user terminal used by the user, determines clear object specification information with a higher value relative to the user according to specification information of the clear object returned by the user terminal, determines a corresponding clear object according to the clear object specification definition, and redetermines the reference image.
In some other embodiments of the present application, to improve the quality of the acquired reference image, a first similarity score of the non-distinct object and the object of the same class, and a second similarity score between the image to be identified and each corresponding distinct object included in the reference image may also be generated, so as to determine the reliability of the reference image according to the weighted results of the first similarity score and the second similarity score.
Referring to fig. 3, fig. 3 is a flowchart of another image recognition method according to an embodiment of the present application, wherein the flowchart 300 includes the following steps:
in step 301, a non-sharp object and a sharp object in an image to be identified are determined.
Step 302, obtaining a reference image of the same class object containing the non-distinct object and the distinct object.
In step 303, the same object is determined as the same object in the same category as the non-distinct object in response to the first object group having the same positional relationship in the image to be identified as the second object group in the reference image.
The above steps 301-303 are identical to the steps 201-203 shown in fig. 2, and the same parts are referred to the corresponding parts of the previous embodiment, and will not be described herein again.
And step 304, in response to the same object as the same class object and the same object as the non-clear object, connecting the image to be identified and the reference image to generate a communication relation diagram.
In this embodiment, when it is determined that the unclear object exists in both the reference image and the image to be identified, the communication relationship between the image to be identified and the candidate image is correspondingly generated, and each image having the communication relationship in the communication relationship graph includes the same unclear object.
Further, after the communication relation between the image to be identified and the reference image is generated, the reference image can be used as a new image to be identified to continuously search other reference images with the relation with the new image to be identified, and a communication relation diagram is determined according to the generated communication relation so as to acquire more reference images with the communication relation.
And step 305, clustering objects in each image forming the communication relation graph to obtain a clustering result.
In this embodiment, the same non-distinct objects included in each image on a specific communication relationship line are determined according to the communication relationship graph generated in the step 304, and each image on the specific communication relationship line is clustered according to the content of the non-distinct objects, so as to obtain a corresponding clustering result.
And step 306, adding the clustering result to the image to be identified as the identification auxiliary information set of the unclear object.
In this embodiment, when the non-distinct objects are used as the clustering center to obtain the clustering result, identification auxiliary information such as the mark information, the image set connection and other forms can be generated for the non-distinct objects in the image to be identified, and the form of the identification auxiliary information is not limited to the forms of adding the content tag, importing the image set and the like, so that the non-distinct objects contained in the image to be identified can be identified according to the identification auxiliary information, and a reference image containing the identifiable high-definition non-distinct objects can be provided for a user to refer to.
Based on the technical scheme provided in the previous embodiment, the embodiment further clusters the images containing the clear objects based on the connected relation, so as to facilitate the auxiliary identification of the non-clear objects in the images to be identified based on the identifiable and high-definition non-clear object content contained in the reference image, and realize the image clustering work based on the non-clear objects, thereby solving the technical problems of low image clustering efficiency, poor effect and incapability of image clustering caused by low image identification precision or reasons such as blocked, uneven angle, unclear shooting and the like in the prior art.
Based on the previous embodiment, the communication relationship route in the communication relationship graph can be readjusted according to the clustering center determined by the clustering result; and generating a new communication relation diagram according to the adjusted communication relation route. In order to further improve the clustering quality, images on the communication relation lines in the communication relation graph are detected, and when the fact that error candidate images which do not contain corresponding non-clear objects exist in the images on the same communication relation line is determined, the error candidate images are adjusted from the communication relation graph so as to ensure that the images on the same communication relation line in the communication relation graph meet requirements, namely the images on the same communication relation line belong to the same clustering center.
In practice, the obtained error candidate images may also contain the same unclear objects and clear objects, so that the content of each error candidate image can be identified to determine a new cluster center, clustering aiming at different cluster centers is realized, the quality and efficiency of image identification and image clustering are improved, and in the process of updating and rejecting the error candidate image from the communication relation graph, the updated error candidate graph can be clustered into a new cluster image set based on the content contained in the error candidate graph, so that the efficiency of image clustering is improved.
On the basis of any of the above embodiments, in combination with the identification requirement of the content of the image to be identified, that is, the object with higher value in the image to be identified, it may be predetermined that when the image to be identified includes at least one of a face object, a body object and a vehicle object whose definition does not meet the preset requirement, the corresponding non-clear object is determined to be at least one of a face object, a body object and a vehicle object, and at least one of license plate information, shooting time information, shooting location information, face vector information and body vector information, which are corresponding to the image to be identified and have higher relevance to the non-clear object, is determined as the clear object, so as to improve quality and efficiency of image identification, and provide an identification result with higher value.
For further understanding, the present application further provides a specific implementation scheme in combination with a specific application scenario, where the image a to be identified (refer to fig. 4-1) includes a non-distinct object a, and the content of the non-distinct object a is expected to be identified, where the specific process is as follows:
and determining a non-clear object a contained in the image A to be identified, and determining a clear object b with the definition larger than that of the non-clear object a from the image A to be identified based on the definition of the non-clear object a.
Multiple images are recalled from the clear object, and a reference image B (see fig. 4-2) containing both the non-clear object homogeneous object a1 and the clear object B is determined therefrom.
The positional relationship of the first object group composed of the non-clear object a and the clear object B in the image a to be recognized (the spatial distance between the non-clear object a and the clear object B is 3 meters, for example) is determined in the image a to be recognized, and the positional relationship of the second object group composed of the same-class object a1 and the clear object B in the reference image B is determined in the reference image B (the spatial distance between the same-class object a1 and the clear object B is 2.9 meters, for example).
And determining that the non-clear object a and the same-category object a1 contained in the image to be identified A and the reference image B are the same object in response to the fact that the position relation of the first object group in the image to be identified A is the same as the position relation of the second object group in the reference image B.
The method and the device determine the reference image based on the clear objects of the non-clear objects contained in the image to be identified, so that the non-clear objects in the image to be identified can be identified in an auxiliary mode through the high-definition content contained in the reference image, the problem that the non-clear objects in the image to be identified cannot be identified due to the fact that the non-clear objects in the image to be identified are blocked, the angles are not correct, the shooting is not clear and the like is solved, and therefore the identification accuracy and the identification efficiency of image identification are improved.
With further reference to fig. 5, as an implementation of the method shown in the foregoing figures, the present application provides an embodiment of an image recognition apparatus, where the embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 2, and the apparatus is specifically applicable to various electronic devices.
As shown in fig. 5, the image recognition apparatus 500 of the present embodiment may include: an object determining unit 501, a reference image acquiring unit 502, and a content discriminating unit 503. Wherein the object determining unit 501 is configured to determine an unclear object and a clear object in an image to be recognized; wherein the definition of the non-clear object is less than the definition of the clear object; a reference image acquisition unit 502 configured to acquire a reference image of the same-class object containing the non-sharp object and the sharp object; wherein the definition of the same class object is greater than the definition of the non-clear object; a content discriminating unit 503 configured to determine that the same-category object and the non-clear object are the same object in response to a positional relationship of the first object group in the image to be recognized being the same as a positional relationship of the second object group in the reference image; wherein the first object group is composed of the non-distinct object and the distinct object, and the second object group is composed of the same class object and the distinct object.
In the present embodiment, in the image recognition apparatus 500: the specific processes of the object determining unit 501, the reference image acquiring unit 502, and the content discriminating unit 503 and the technical effects thereof may refer to the relevant descriptions of steps 201 to 203 in the corresponding embodiment of fig. 2, and are not repeated herein.
In some optional implementations of the present embodiment, the object determining unit 501 includes: the image to be identified acquisition subunit is configured to acquire an image to be identified uploaded by a user through the user terminal; an instruction information acquisition subunit configured to acquire object instruction information of the user for the image to be identified through the user terminal; a non-clear object determination subunit configured to determine a non-clear object according to the object indication information; and a clear object determination subunit configured to determine an object having a definition greater than the non-definition in the image to be recognized as a clear object.
In some optional implementations of the present embodiment, the image recognition apparatus 500 further includes: a reference image pushing unit configured to push the reference image to the user terminal in response to the reference image including more than a preset number of the distinct objects; and the reference image updating unit is configured to redetermine the reference image according to the clear object specification information returned by the user terminal.
In some optional implementations of this embodiment, the non-distinct object includes at least one of a face object, a body object, and a vehicle object, and the distinct object includes at least one of license plate information, shooting time information, shooting location information, face vector information, and body vector information.
In some optional implementations of the present embodiment, the image recognition apparatus 500 further includes: a communication relation diagram generating unit configured to generate a communication relation diagram in response to the same-category object and the non-clear object being the same object, connecting the image to be identified and the reference image; the object clustering unit is configured to cluster objects in each image forming the communication relation graph to obtain a clustering result; and an auxiliary information generating unit configured to add the clustering result as a set of identification auxiliary information of the non-sharp object to the image to be identified.
In some optional implementations of the present embodiment, the image recognition apparatus 500 further includes: the communication relation diagram updating unit is configured to readjust the communication relation route in the communication relation diagram according to the clustering center determined by the clustering result; and the communication relation diagram generating unit is further configured to generate a new communication relation diagram according to the adjusted communication relation route.
The embodiment exists as an embodiment of the apparatus corresponding to the embodiment of the method, and the image recognition apparatus provided in the embodiment determines the reference image based on the clear object of the non-clear object included in the image to be recognized, so as to facilitate the auxiliary recognition of the non-clear object in the image to be recognized through the high-definition content included in the reference image, solve the problem that the non-clear object in the image to be recognized cannot be recognized due to reasons such as shielding, incorrect angle, unclear shooting, and the like, and improve the recognition accuracy and the recognition efficiency of the image recognition.
According to embodiments of the present application, there is also provided an electronic device, a readable storage medium and a computer program product.
Fig. 6 shows a schematic block diagram of an example electronic device 600 that may be used to implement embodiments of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the application described and/or claimed herein.
As shown in fig. 6, the apparatus 600 includes a computing unit 601 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 602 or a computer program loaded from a storage unit 608 into a Random Access Memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 may also be stored. The computing unit 601, ROM 602, and RAM 603 are connected to each other by a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
Various components in the device 600 are connected to the I/O interface 605, including: an input unit 606 such as a keyboard, mouse, etc.; an output unit 607 such as various types of displays, speakers, and the like; a storage unit 608, such as a magnetic disk, optical disk, or the like; and a communication unit 609 such as a network card, modem, wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The computing unit 601 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 601 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the respective methods and processes described above, such as an image recognition method. For example, in some embodiments, the image recognition method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 600 via the ROM 602 and/or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the image recognition method described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform the image recognition method by any other suitable means (e.g. by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be a special or general purpose programmable processor, operable to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for carrying out methods of the present application may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this application, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.
The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so as to solve the defects of large management difficulty and weak service expansibility in the traditional physical host and virtual private server (VPS, virtual Private Server) service. Servers may also be divided into servers of a distributed system or servers that incorporate blockchains.
According to the technical scheme, the reference image is determined based on the clear objects of the non-clear objects contained in the image to be identified, so that the non-clear objects in the image to be identified can be identified in an auxiliary mode through the high-definition content contained in the reference image, the problem that the non-clear objects in the image to be identified cannot be identified due to the fact that the non-clear objects in the image to be identified are blocked, the angle is not correct, the shooting is not clear and the like is solved, and therefore the identification accuracy and the identification efficiency of image identification are improved.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present application may be performed in parallel, sequentially, or in a different order, provided that the desired results of the technical solutions disclosed in the present application can be achieved, and are not limited herein.
The above embodiments do not limit the scope of the application. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application are intended to be included within the scope of the present application.
Claims (12)
1. An image recognition method, comprising:
determining an unclear object and a clear object in an image to be identified; wherein the definition of the non-distinct object is less than the definition of the distinct object;
acquiring a reference image of the same-class object containing the non-clear object and the clear object; wherein the definition of the same class object is greater than the definition of the non-clear object;
determining that the same-category object and the non-clear object are the same object in response to the fact that the position relation of the first object group in the image to be identified is the same as the position relation of the second object group in the reference image; wherein the first object group consists of the non-distinct objects and the distinct objects, and the second object group consists of the same class objects and the distinct objects;
responding to the same object as the non-clear object, connecting the image to be identified with the reference image, and generating a communication relation diagram;
clustering objects in each image forming the communication relation graph to obtain a clustering result;
and adding the clustering result to the image to be identified as an identification auxiliary information set of the non-clear object.
2. The method of claim 1, wherein the determining non-sharp objects and sharp objects in the image to be identified comprises:
acquiring an image to be identified and object indication information aiming at the image to be identified, which are uploaded by a user through a user terminal;
and determining an unclear object according to the object indication information, and determining an object with definition greater than that of the unclear object in the image to be identified as a clear object.
3. The method of claim 2, further comprising:
pushing the reference image to the user terminal in response to the reference image containing more than a preset number of the clear objects;
and re-determining the reference image according to the clear object specification information returned by the user terminal.
4. The method of claim 1, the non-distinct object comprising at least one of a face object, a body object, and a vehicle object, the distinct object comprising at least one of license plate information, shooting time information, shooting location information, face vector information, and body vector information.
5. The method of claim 1, further comprising:
readjusting the communication relation route in the communication relation graph according to the clustering center determined by the clustering result;
And generating a new communication relation diagram according to the adjusted communication relation route.
6. An image recognition apparatus comprising:
an object determining unit configured to determine an unclear object and a clear object in an image to be recognized; wherein the definition of the non-distinct object is less than the definition of the distinct object;
a reference image acquisition unit configured to acquire a reference image of a same-class object containing the non-clear object and the clear object; wherein the definition of the same class object is greater than the definition of the non-clear object;
a content discriminating unit configured to determine that the same-category object and the non-clear object are the same object in response to a positional relationship of a first object group in the image to be identified being the same as a positional relationship of a second object group in the reference image; wherein the first object group consists of the non-distinct objects and the distinct objects, and the second object group consists of the same class objects and the distinct objects;
a communication relation diagram generating unit configured to generate a communication relation diagram in response to the same class object and the non-clear object being the same object, connecting the image to be identified and the reference image;
The object clustering unit is configured to cluster objects in each image forming the communication relation graph to obtain a clustering result;
and an auxiliary information generating unit configured to add the clustering result as a recognition auxiliary information set of the unclear object to the image to be recognized.
7. The apparatus of claim 6, wherein the object determination unit comprises:
the image to be identified acquisition subunit is configured to acquire an image to be identified uploaded by a user through the user terminal;
an instruction information acquisition subunit configured to acquire object instruction information of the user for the image to be identified through the user terminal;
a non-clear object determination subunit configured to determine a non-clear object according to the object indication information;
and a clear object determination subunit configured to determine an object with a definition greater than that of the non-clear object in the image to be recognized as a clear object.
8. The apparatus of claim 7, further comprising:
a reference image pushing unit configured to push the reference image to the user terminal in response to the reference image containing more than a preset number of the distinct objects;
And the reference image updating unit is configured to redetermine the reference image according to the clear object specification information returned by the user terminal.
9. The apparatus of claim 6, the non-distinct object comprising at least one of a face object, a body object, and a vehicle object, the distinct object comprising at least one of license plate information, shooting time information, shooting location information, face vector information, and body vector information.
10. The apparatus of claim 6, further comprising:
the communication relation diagram updating unit is configured to readjust the communication relation route in the communication relation diagram according to the clustering center determined by the clustering result; and
the communication relation diagram generating unit is further configured to generate a new communication relation diagram according to the adjusted communication relation route.
11. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the image recognition method of any one of claims 1-5.
12. A non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the image recognition method of any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110245509.2A CN112966606B (en) | 2021-03-05 | 2021-03-05 | Image recognition method, related device and computer program product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110245509.2A CN112966606B (en) | 2021-03-05 | 2021-03-05 | Image recognition method, related device and computer program product |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112966606A CN112966606A (en) | 2021-06-15 |
| CN112966606B true CN112966606B (en) | 2023-07-21 |
Family
ID=76276661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110245509.2A Active CN112966606B (en) | 2021-03-05 | 2021-03-05 | Image recognition method, related device and computer program product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112966606B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930271A (en) * | 2012-09-21 | 2013-02-13 | 博康智能网络科技股份有限公司 | Method for identifying taxicabs in real time by utilizing video images |
| US9036083B1 (en) * | 2014-05-28 | 2015-05-19 | Gracenote, Inc. | Text detection in video |
| CN109934232A (en) * | 2019-02-01 | 2019-06-25 | 中山安信通机器人制造有限公司 | Vehicle identification method, computer installation and computer readable storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104596929B (en) * | 2013-10-31 | 2017-06-23 | 国际商业机器公司 | Determine the method and apparatus of air quality |
-
2021
- 2021-03-05 CN CN202110245509.2A patent/CN112966606B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930271A (en) * | 2012-09-21 | 2013-02-13 | 博康智能网络科技股份有限公司 | Method for identifying taxicabs in real time by utilizing video images |
| US9036083B1 (en) * | 2014-05-28 | 2015-05-19 | Gracenote, Inc. | Text detection in video |
| CN109934232A (en) * | 2019-02-01 | 2019-06-25 | 中山安信通机器人制造有限公司 | Vehicle identification method, computer installation and computer readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 低清晰度车牌识别技术的研究与应用;孙阳;王云龙;何文铎;;华中科技大学学报(自然科学版)(S1);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112966606A (en) | 2021-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3852008A2 (en) | Image detection method and apparatus, device, storage medium and computer program product | |
| JP7393472B2 (en) | Display scene recognition method, device, electronic device, storage medium and computer program | |
| CN113222942A (en) | Training method of multi-label classification model and method for predicting labels | |
| CN113360700B (en) | Training of image-text retrieval model, image-text retrieval method, device, equipment and medium | |
| CN113392794B (en) | Vehicle line crossing identification method and device, electronic equipment and storage medium | |
| CN112860993A (en) | Method, device, equipment, storage medium and program product for classifying points of interest | |
| CN113378855A (en) | Method for processing multitask, related device and computer program product | |
| CN113378958A (en) | Automatic labeling method, device, equipment, storage medium and computer program product | |
| US20230005171A1 (en) | Visual positioning method, related apparatus and computer program product | |
| CN113887615A (en) | Image processing method, apparatus, device and medium | |
| CN114218931A (en) | Information extraction method and device, electronic equipment and readable storage medium | |
| CN115690443A (en) | Feature extraction model training method, image classification method and related device | |
| CN113033373B (en) | Method for training face recognition model and recognizing face and related device | |
| CN113326766A (en) | Training method and device of text detection model and text detection method and device | |
| CN113591580A (en) | Image annotation method and device, electronic equipment and storage medium | |
| CN112966606B (en) | Image recognition method, related device and computer program product | |
| US20220351495A1 (en) | Method for matching image feature point, electronic device and storage medium | |
| CN114429631B (en) | Three-dimensional object detection method, device, equipment and storage medium | |
| CN113627526B (en) | Vehicle identification recognition method and device, electronic equipment and medium | |
| CN114444514B (en) | Semantic matching model training method, semantic matching method and related device | |
| CN114329164B (en) | Method, apparatus, device, medium, and article for processing data | |
| CN114639143A (en) | Portrait filing method, equipment and storage medium based on artificial intelligence | |
| CN114398434A (en) | Structured information extraction method and device, electronic equipment and storage medium | |
| CN113870428A (en) | Scene map generation method, related device and computer program product | |
| US20220383626A1 (en) | Image processing method, model training method, relevant devices and electronic device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |