Image Processing Method, Device, Equipment, Medium and Product for Object Attribute Recognition

Abstract

An embodiment of the present invention provides an image processing method applied to object attribute recognition, which can be applied to the field of artificial intelligence technology. The image processing method applied to object attribute recognition includes: obtaining a subgraph to be recognized of a target recognition image; determining an attribute recognition label of the subgraph to be recognized according to the matching relationship between the target feature value of the subgraph to be recognized and the set of template feature values of a preset matching template; and determining corresponding object attribute information according to the attribute recognition label. An embodiment of the present invention also provides an image processing device, equipment, storage medium and program product applied to object attribute recognition.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, specifically to the field of image processing technology, and more specifically to an image processing method, device, equipment, medium and product for object attribute recognition. Background Art

[0002] Artificial Intelligence (AI for short) is an important driving force for the new round of scientific and technological revolution and industrial transformation. It is a new key technology science that studies, develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. As an important part of intelligent science, artificial intelligence attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a way similar to human intelligence. Artificial intelligence is a very broad science, including robots, speech recognition, image recognition, natural language processing, expert systems, machine learning, and computer vision, etc.

[0003] Computer Vision (CV for short) is a key branch of artificial intelligence, aiming to enable a computer to understand and process visual information, that is, "see". Computer vision involves many tasks, including image classification, object detection, image segmentation, pose estimation, object tracking, image generation, etc. Its ultimate goal is to enable a machine to accurately understand visual data in different scenarios and make intelligent responses and decisions. And the premise of understanding visual data is that the computer can accurately identify the objects in the acquired images, specifically judged by the accuracy of object attribute recognition. However, the existing object attribute recognition technologies usually can only achieve relatively accurate recognition for a certain specific attribute feature and cannot effectively support the accurate recognition of multiple or even all necessary attribute features. Summary of the Invention

[0004] In view of at least one of the above problems, the embodiments of the present invention aim to provide an image processing method, device, equipment, medium and product for object attribute recognition that can take into account the accurate recognition of multiple object attribute features, thereby providing a technical solution that can be extended to the accurate recognition of more than 2 object attribute features at a lower cost, in order to achieve a higher level of intelligence.

[0005] One aspect of the embodiments of the present invention provides an image processing method for object attribute recognition, which includes: obtaining a sub-image to be recognized of a target recognition image; determining an attribute recognition label of the sub-image to be recognized according to the matching relationship between the target feature value of the sub-image to be recognized and the set of template feature values of a preset matching template; and determining corresponding object attribute information according to the attribute recognition label.

[0006] According to an embodiment of the present invention, in obtaining a sub - image to be recognized of a target recognition image, it includes: obtaining mask data information of the target recognition image; and obtaining the sub - image to be recognized of the target recognition image according to the mask data information.

[0007] According to an embodiment of the present invention, in obtaining the mask data information of the target recognition image, it includes: obtaining a category sub - image of the target recognition image that matches preset category information; and extracting the mask data information of the category sub - image.

[0008] According to an embodiment of the present invention, in obtaining the sub - image to be recognized of the target recognition image according to the mask data information, it includes: generating a background image corresponding to the mask data information; and obtaining the sub - image to be recognized according to the background image.

[0009] According to an embodiment of the present invention, in determining the attribute recognition label of the sub - image to be recognized according to the matching relationship between the target feature value of the sub - image to be recognized and the set of template feature values of a preset matching template, it includes: obtaining the similarity between the target feature value and each template feature value in the set of template feature values; and determining the attribute recognition label of the sub - image to be recognized according to the similarity.

[0010] According to an embodiment of the present invention, in determining the attribute recognition label of the sub - image to be recognized according to the similarity, it includes: when the similarity is greater than or equal to a preset threshold, determining the attribute recognition label of the template feature value corresponding to the similarity.

[0011] According to an embodiment of the present invention, in determining the corresponding object attribute information according to the attribute recognition label, it includes: sorting the attribute recognition labels according to the similarity; determining the object attribute information corresponding to the attribute recognition label after sorting; or determining the object attribute information corresponding to the attribute recognition label with the maximum similarity.

[0012] According to an embodiment of the present invention, the image - processing method applied to object attribute recognition further includes: updating the corresponding attribute recognition label according to the dynamic detection result of a preset matching template.

[0013] Another aspect of the embodiment of the present invention provides an image - processing device applied to object attribute recognition, which includes a sub - image acquisition module, a label determination module, and an information determination module. The sub - image acquisition module is used to obtain the sub - image to be recognized of the target recognition image; the label determination module is used to determine the attribute recognition label of the sub - image to be recognized according to the matching relationship between the target feature value of the sub - image to be recognized and the set of template feature values of a preset matching template; and the information determination module is used to determine the corresponding object attribute information according to the attribute recognition label.

[0014] Another aspect of an embodiment of the present invention provides an electronic device, including one or more processors and a memory for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors are caused to execute the above-mentioned image processing method applied to object attribute recognition.

[0015] Another aspect of an embodiment of the present invention provides a computer-readable storage medium, on which executable instructions are stored. When the instructions are executed by a processor, the processor is caused to execute the above-mentioned image processing method applied to object attribute recognition.

[0016] Another aspect of an embodiment of the present invention provides a computer program product, including a computer program which, when executed by a processor, implements the above-mentioned image processing method applied to object attribute recognition.

[0017] The image processing method applied to object attribute recognition provided by an embodiment of the present invention can at least partially solve the problems in the related art such as the inability to expand object attribute recognition at low cost, and thus can at least achieve one of the following technical effects:

[0018] The image processing method applied to object attribute recognition provided by an embodiment of the present invention can meet the high-precision recognition of more than two object attribute features by means of the graphic matching strategy defined by the matching relationship between the target feature value of the sub-graph to be recognized and the set of template feature values of the preset matching template. For example, color, shape, texture, material, etc. can all be accurately recognized in parallel. At the same time, during the image processing process, background noise can be effectively reduced, greatly improving the accuracy of object attribute recognition. Moreover, for the newly added attribute features in the preset matching template, recognition can also be directly achieved without performing repetitive work such as iterative tuning, and the attribute categories can be arbitrarily expanded, with stronger versatility. Further, combined with the hot loading mechanism, online update of the newly added attribute categories can be realized at low cost. Therefore, the above-mentioned image processing method of the embodiment of the present invention can meet the requirement of improving the recognition versatility of the whole system at low cost, has higher recognition accuracy, and thus greatly improves the intelligent level of recognition.

[0019] It should be understood that the above general description and the following specific embodiments are only exemplary and explanatory, and cannot limit the scope claimed by the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Through the following description of the embodiments of the present invention with reference to the drawings, the above content and other objects, features and advantages of the present invention will become clearer. In the drawings:

[0021] Figure 1Schematically shows an application scenario diagram of an image processing method, apparatus, device, medium, and program product for object attribute recognition according to an embodiment of the present invention;

[0022] Figure 2 Schematically shows a flowchart of an image processing method for object attribute recognition according to an embodiment of the present invention;

[0023] Figure 3A Schematically shows an acquisition scenario diagram of a category sub - diagram of an image processing method for object attribute recognition according to an embodiment of the present invention;

[0024] Figure 3B Schematically shows an extraction scenario diagram of mask data information of a category sub - diagram of an image processing method for object attribute recognition according to an embodiment of the present invention;

[0025] Figure 4 Schematically shows a picture template diagram of an image processing method for object attribute recognition according to an embodiment of the present invention;

[0026] Figure 5 Schematically shows a structural block diagram of an image processing apparatus for object attribute recognition according to an embodiment of the present invention; and

[0027] Figure 6 Schematically shows a block diagram of an electronic device suitable for implementing an image processing method for object attribute recognition according to an embodiment of the present invention.

[0028] The above - mentioned drawings are a part of the specification of the embodiments of the present invention, which illustrate the exemplary embodiments of the present invention. The attached drawings, together with the description of the specification, are used to explain the principles of the embodiments of the present invention. It should be understood that the above general description of the drawings and the following specific embodiments are only exemplary and explanatory, and they do not limit the scope of what the present invention intends to claim. Detailed Embodiments

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer and more understandable, the following will clearly explain the spirit of what is disclosed in the present invention with reference to the drawings and detailed descriptions. Any person skilled in the relevant technical field, after understanding the embodiments of the content of the present invention, can make changes and modifications based on the techniques taught by the content of the present invention, without departing from the spirit and scope of the content of the present invention.

[0030] The exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but do not limit the present invention. Additionally, elements / components using the same or similar reference numerals in the drawings and embodiments are used to represent the same or similar parts.

[0031] Regarding "first", "second",... etc. used in the present invention, they do not particularly refer to the meaning of order or sequence, nor are they used to limit the present invention. They are only used to distinguish elements or operations described with the same technical terms.

[0032] Regarding the directional terms used in the present invention, such as: up, down, left, right, front or back, etc., they are only references to the directions in the attached drawings. Therefore, the directional terms used are for illustration and not for limitation of the creation.

[0033] Regarding "comprising", "including", "having", "containing", etc. used in the present invention, they are all open-ended terms, that is, they mean including but not limited to.

[0034] Regarding "and / or" used in the present invention, it includes any one or all combinations of the described things.

[0035] Regarding "a plurality of" in the present invention, it includes "two" and "more than two"; regarding "a plurality of groups" in the present invention, it includes "two groups" and "more than two groups".

[0036] Regarding the terms "substantially", "about", etc. used in the present invention, they are used to modify any quantity or error that can vary slightly, but these slight variations or errors do not change their essence. Generally speaking, the range of such slight variations or errors modified by such terms can be 20% in some embodiments, 10% in some embodiments, 5% in some embodiments or other values. Those skilled in the art should understand that the aforementioned values can be adjusted according to actual needs and are not limited thereto.

[0037] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used here should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0038] In the case of using expressions such as "at least one of A, B, and C, etc.", generally, it should be interpreted according to the meaning that those skilled in the art usually understand this expression (for example, "a system having at least one of A, B, and C" should include, but not be limited to, a system having only A, only B, only C, having A and B, having A and C, having B and C, and / or having A, B, and C, etc.). In the case of using expressions such as "at least one of A, B, or C, etc.", generally, it should be interpreted according to the meaning that those skilled in the art usually understand this expression (for example, "a system having at least one of A, B, or C" should include, but not be limited to, a system having only A, only B, only C, having A and B, having A and C, having B and C, and / or having A, B, and C, etc.). Those skilled in the art should also understand that substantially any disjunctive conjunctions and / or phrases representing two or more alternative items, whether in the specification, claims, or drawings, should be understood to give the possibility of including one of these items, either side of these items, or both items. For example, the phrase "A or B" should be understood to include the possibility of "A" or "B", or "A and B".

[0039] In the field of artificial intelligence technology, image recognition can identify various types of information from the acquired images, such as spatial information (spatial size, orientation, and even room type, etc.), object information (object category, size, position, orientation, color, and shape, etc.), and human information (2D skeleton points, 3D skeleton points, actions, and face ID, etc.). Among them, accurately identifying various attribute information of objects in images through computer vision technology has great practical value for industrial applications in various fields. For example, in the agricultural field, the growth status can be judged by identifying the color of plants; in the logistics field, the loading method can be determined by the shape of the express delivery, etc.

[0040] Currently, the mainstream object attribute recognition technologies mainly perform classification tasks based on the RGB or HSV information of the target area in the image. The classification methods are usually based on neural networks or template matching. However, these traditional methods require a large amount of human and financial resources in the collection and annotation of training data and template design, resulting in a high input cost for object attribute recognition technologies. Moreover, the existing technologies still lack the batch recognition of multi-attribute features. Each attribute usually needs to be implemented based on a single model, and there is interference between the recognition of different attributes. It is difficult to perform batch and continuous recognition for a specific attribute without affecting the recognition of other attributes. Moreover, the model output is usually relatively fixed, and the recognition ability is relatively fixed, and it is impossible to introduce the recognition of new attributes. In short, the current object attribute recognition technologies mainly suffer from the lack of a general object attribute recognition model, that is, there is very little ability to support the recognition of multiple attributes such as color, shape, texture, and material with high precision, and it is also difficult to expand the recognition of new attribute categories in a low-cost manner.

[0041] Furthermore, for the recognition of a certain object attribute in an image, different recognition abilities are required in different scenarios. For example, in a specific business scenario, it may be necessary to recognize red and yellow colors, while in other scenarios, it may be necessary to recognize pink and green colors. In traditional technologies, for such the same type of recognition tasks in different scenarios, from the realization of red and yellow color recognition to pink and green color recognition, it is still necessary to spend manpower on similar repetitive work, such as collecting training data, cleaning data, data augmentation, training models, and iterative tuning, etc. This repetitive work greatly increases the R & D cost. Therefore, it is necessary to provide a more intelligent recognition solution that can quickly customize recognition requirements, update recognition requirements at any time, and even expand to the recognition ability of other object attributes except color, such as shape, material, texture, etc., so as to greatly improve the intelligence level of the intelligent agent.

[0042] In view of at least one of the technical problems existing in the above-mentioned prior art, the embodiments of the present invention aim to provide an image processing method, device, equipment, medium, and product for object attribute recognition that can take into account the accurate recognition of multiple object attribute features, thereby providing a technical solution that can quickly customize recognition requirements, update recognition requirements at any time, and even expand to the recognition ability of more other object attributes, so as to greatly improve the intelligence level of the intelligent agent and expand the accurate recognition of more than 2 object attribute features at a low cost.

[0043] One aspect of an embodiment of the present invention provides an image processing method applied to object attribute recognition, which includes: obtaining a sub-image to be recognized of a target recognition image; determining an attribute recognition label of the sub-image to be recognized according to the matching relationship between the target feature value of the sub-image to be recognized and the set of template feature values of a preset matching template; and determining corresponding object attribute information according to the attribute recognition label.

[0044] Figure 1 Schematically shows an application scenario diagram of an image processing method, device, equipment, medium and program product applied to object attribute recognition according to an embodiment of the present invention.

[0045] As Figure 1 shown, the application scenario 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used to provide a medium for 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, etc.

[0046] Users can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications may be installed on the terminal devices 101, 102, 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0047] The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, and desktop computers, etc.

[0048] The server 105 may be a server providing various services, such as a background management server that supports websites browsed by users using the terminal devices 101, 102, 103 (only for example). The background management server may analyze and process data such as received user requests, and feedback the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0049] It should be noted that the image processing method for object attribute recognition provided by the embodiments of the present invention can generally be executed by the server 105. Correspondingly, the image processing device for object attribute recognition provided by the embodiments of the present invention can generally be disposed in the server 105. The image processing method for object attribute recognition provided by the embodiments of the present invention can also be executed by a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, 103, and / or the server 105. Correspondingly, the image processing device for object attribute recognition provided by the embodiments of the present invention can also be disposed in a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, 103, and / or the server 105.

[0050] It should be understood that Figure 1 the numbers of terminal devices, networks, and servers in

[0051] are merely illustrative. According to the implementation requirements, there can be any number of terminal devices, networks, and servers. Figure 1 Based on the Figures 2 to 4 scenario described below, the image processing method for object attribute recognition of the disclosed embodiments will be described in detail through

[0052] As Figure 2 shown, one aspect of the embodiments of the present invention provides an image processing method for object attribute recognition, which includes operations S201 to S203.

[0053] In operation S201, a sub - graph to be recognized of the target recognition image is obtained;

[0054] In operation S202, according to the matching relationship between the target feature value of the sub - graph to be recognized and the set of template feature values of the preset matching template, the attribute recognition label of the sub - graph to be recognized is determined; and

[0055] In operation S203, the corresponding object attribute information is determined according to the attribute recognition label.

[0056] The target recognition image is an image with a target recognition object, which is usually related to the actual recognition requirements of the intelligent agent. For example, when an intelligent vehicle is driving on the road, it needs to recognize the objects (vehicles, people, animals, etc.) in all directions (front, back, left, and right) during the driving process. The images continuously captured by the camera set on the vehicle itself can be used as the target recognition image, and the target recognition object is in the target recognition image. Only by accurately recognizing the attribute features of the target recognition object can the intelligent agent use it as the basis for executing the next action.

[0057] The sub - graph to be recognized is a local processed image of the target recognition object on the target recognition image. The local processed image can exclude the recognition of non - target recognition objects (such as background images) as much as possible, improve the recognition accuracy of the corresponding object attributes of the target recognition object, and at the same time avoid unnecessary recognition. For example, if only the people in the target recognition image need to be recognized, only the local image of the framed people needs to be processed to generate the sub - graph to be recognized.

[0058] The target feature value is the feature value of the sub - graph to be recognized related to the target recognition attribute corresponding to the target recognition object. For example, if the target recognition attribute is color, the target feature value can be the feature value related to color. In addition, the target recognition attribute can also be multiple categories such as color, shape, and material at the same time. At this time, the target feature value can be the feature value related to multiple categories such as color, shape, and material, rather than the feature value of a single attribute.

[0059] The preset matching template is template data with corresponding attribute recognition labels. These template data can define the corresponding feature values of each attribute feature, and can be specifically embodied in the form of text (such as text.txt) and / or graphics and text (such as picture.jpg).

[0060] For example, the text in txt format of "a red object with white background" can be used as the attribute recognition label for the color "red", which can be used to define the object attribute color as "red". Correspondingly, the picture in picture format of "pink.jpg" can be used as the attribute recognition picture for the color "pink", and the naming text of this picture can be used as the attribute recognition label for this color attribute. Among them, when these texts in txt format are presented in the form of a list, they can form a text template list in txt format, and this text template list can be used as a component of this preset matching template. At the same time, these attribute pictures in jpg format with text labels can be combined to form another component of this preset matching template.

[0061] Therefore, when extracting the corresponding text feature values or image feature values from the text of these text template lists or a certain attribute picture alone, the corresponding template feature values can be formed to form a template feature value set (i.e., a set of feature values).

[0062] The template eigenvalue set can include all the eigenvalues of the preset matching templates for text and picture templates, that is, it can include the matching eigenvalues of various attributes. These matching eigenvalues can establish a mapping relationship with the target eigenvalues of the sub-graph to be recognized according to the preset matching relationship. Based on this kind of mapping relationship, the template eigenvalue that matches the target eigenvalue can be found from the template eigenvalue set, and the attribute recognition label corresponding to this template eigenvalue can correspond to the corresponding object attribute. Therefore, high-precision parallel recognition of multi-attribute categories can be achieved. The processes of each attribute recognition are carried out in parallel, which can improve the recognition efficiency while ensuring the recognition accuracy, and improve the recognition versatility.

[0063] The attribute recognition label (Label) is the preset text or picture information directly related to the attribute in the preset matching template, and it can correspond to the eigenvalue of the corresponding attribute. For example, the text in txt format of "a red object with white background" can be used as the attribute recognition label for the color "red", which can be used to define the object attribute color as "red". This text has the corresponding text eigenvalue B1 that can correspond to this attribute recognition label of "red". Correspondingly, the picture in picture format of "pink.jpg" can be used as the attribute recognition picture for the color "pink". The named text "pink.jpg" of this picture can be used as the attribute recognition label for this color attribute, and the picture eigenvalue B2 of this picture can correspond to this attribute recognition label of "pink".

[0064] Therefore, the corresponding object attribute information can be directly defined through the attribute recognition label. This object attribute information can be the text name, eigenvalue, etc. of the object attribute, which can be specifically reflected in the form of text or a combination of text and pictures. In addition, it can also include the coding information of the corresponding attribute, which can be determined according to the actual attribute recognition requirements.

[0065] The image processing method for object attribute recognition provided by the embodiments of the present invention can meet the parallel high-precision recognition of two or more object attribute features by means of the graphic matching strategy defined by the matching relationship between the target feature values of the sub-graph to be recognized and the set of template feature values of the preset matching template. For example, color, shape, texture, material, etc. can all be accurately recognized in parallel. At the same time, during the image processing process, background noise can be effectively reduced, greatly improving the accuracy of object attribute recognition. Moreover, for the newly added attribute features in the preset matching template, recognition can be directly achieved without repetitive work such as iterative tuning, etc., and the attribute categories can be arbitrarily expanded, with stronger versatility. Further, combined with the hot loading mechanism, online update of the newly added attribute categories can be achieved at low cost. Therefore, the above-mentioned image processing method of the embodiments of the present invention can meet the requirement of improving the recognition versatility of the entire system at low cost, with higher recognition accuracy, thus greatly improving the intelligent level of recognition.

[0066] As Figures 2 - 4 shown, according to an embodiment of the present invention, in the operation of obtaining the sub-graph to be recognized of the target recognition image in S201, it includes:

[0067] Obtain the mask data information of the target recognition image;

[0068] Obtain the sub-graph to be recognized of the target recognition image according to the mask data information.

[0069] The target recognition image can define corresponding mask values (Mask) according to the distribution positions of its corresponding attribute features and other information in the image. These mask values are used to define the range of local operation processing on the target recognition image, so as to realize further extraction, analysis, and fusion processing of specific target areas. The mask data information is the data information related to such mask values, which can define the color, pixels, and position distribution of corresponding points on the image.

[0070] Therefore, through the mask data information, the extraction operation of the sub-graph to be recognized can be more accurately realized, making the acquisition process of the recognition sub-graph faster and more effective, and excluding more background factors. For example, after converting the target recognition image into mask data information, the distribution of the same mask value at different positions in the image can define a certain object. For example, the distribution area of the mask value "1" in the entire target recognition image can be defined as the distribution area of the target recognition object "car". Therefore, the local image where the target recognition object "car" is located can be extracted according to the distribution area of these values "1", and further processed to generate the corresponding sub-graph to be recognized.

[0071] It can be seen that in this way, the noise interference of non-target recognition objects can be effectively excluded, making the recognition of target recognition objects more accurate, thus ensuring the more accurate recognition of the target recognition attributes of target recognition objects.

[0072] As Figures 2 - 4 shown, according to an embodiment of the present invention, in obtaining the mask data information of the target recognition image, it includes:

[0073] Obtain the category sub-image of the target recognition image that matches the preset category information;

[0074] Extract the mask data information of the category sub-image.

[0075] The preset category information can be embodied in the form of text and / or codes, and is specifically used as the extraction basis for the category sub-image of the target recognition image. It can define the target recognition object that meets the user's recognition needs and its corresponding attribute information, which can be specifically realized through user preset, or can be used as the input data of the intelligent agent through instant operation.

[0076] The category sub-image is a local image (i.e., Patch) generated by performing the extraction operation of the target recognition object on the target recognition image according to the preset category information. The specific extraction operation can be implemented by extraction tools such as the open-set object detector (Grounding-DINO).

[0077] As Figure 3A shown, if the target recognition image is the left image and the preset category information is "dog. stick.", then according to the information of "dog. stick.", the extraction operation can be performed by Grounding-DINO, and according to the matching degree defined by the matching threshold, 3 corresponding category sub-images as shown on the right side of the figure are generated, namely 3 local images of "yellow dog, black dog and stick", and the specific details are not elaborated here.

[0078] Furthermore, perform a further extraction operation on the mask data information of the above-mentioned category sub-image, which can be specifically realized by segmentation tools such as an image segmenter (such as Mobilesam). As Figure 3B shown, the category sub-image is a local image of "car" with a partial background image generated after patch extraction. Using Mobilesam to perform a further extraction operation on the mask data information of this category sub-image can accurately extract only the image distribution area (blue area) where the "car" is located, and generate the corresponding mask data information (i.e., Mask).

[0079] Through the extraction process of the category sub-image and the extraction of the mask data information, the original background noise of the image can be effectively reduced, and the high-precision segmentation of the background image can be realized, so as to ensure that all the effective information corresponding to the target recognition object in the category sub-image can be accurately obtained.

[0080] As Figures 2 - 4As shown, according to an embodiment of the present invention, in obtaining a sub - graph to be recognized of a target recognition image based on mask data information, it includes:

[0081] Generate a background image corresponding to the mask data information;

[0082] Obtain the sub - graph to be recognized based on the background image.

[0083] Based on the mask data information extracted from the category sub - graph, a solid - color image with the same size can be constructed as the background image corresponding to the above - mentioned mask data information. This mask data information can define the size information of the target recognition object extracted in the original image. Therefore, based on this size information, a solid - color image can be directly generated as the background image.

[0084] Furthermore, by the difference between the foreground information of the category sub - graph defined by the mask data information and the background information of the background image, the fusion process of the background image and the foreground image of the category sub - graph is realized, constituting a sub - graph to be recognized with a solid - color background.

[0085] Therefore, the sub - graph to be recognized can effectively exclude the influence of background noise, and at the same time can contain all the effective information of the target recognition object, making the subsequent recognition of the target recognition attributes more accurate. Among them, the color of the background image can be matched according to the color of the foreground image of the category sub - graph, not limited to a certain specific color, and will not be elaborated here.

[0086] As Figures 2 - 4 shown, according to an embodiment of the present invention, in determining the attribute recognition label of the sub - graph to be recognized based on the matching relationship between the target feature value of the sub - graph to be recognized and the set of template feature values of the preset matching template in operation S202, it includes:

[0087] Obtain the similarity between the target feature value and each template feature value in the set of template feature values;

[0088] Determine the attribute recognition label of the sub - graph to be recognized according to the similarity.

[0089] Performing feature value extraction on the sub - graph to be recognized can obtain the corresponding multi - dimensional target feature value. The specific extraction operation can be realized through the corresponding feature extraction tool, such as Clip, etc. Among them, the sub - graph to be recognized is an image of the target recognition object with a solid - color background relative to the original target recognition image. For example, through the PictureEncoder of Clip, 512 - dimensional target feature value A of the sub - graph to be recognized can be extracted.

[0090] Each template feature value in the template feature value set of the preset matching template is also a value corresponding to the target feature value in the corresponding dimension. For example, the template feature value set is [B1, B2, B3, ..., Bn], where each Bx value is a template feature value and is also a value of the same dimension (512 dimensions) as the target feature value A.

[0091] Furthermore, the target feature value and each template feature value are matched. Specifically, the cosine similarity can be calculated between the target feature value A and each template feature value Bx in the template feature value set [B1, B2, B3, ..., Bn] respectively to determine the corresponding similarity. According to the magnitudes of these similarities, the template feature value Bx that matches the target feature value A is determined, and then the corresponding attribute category label is determined based on the matching template feature value Bx.

[0092] Specifically, assume that three attributes, namely color, material, and shape, can be used as the target recognition attributes that users are concerned about. After obtaining the corresponding target feature value of the target recognition object, all subdirectories colortemp (color), material temp (material), and shape temp (shape) in the current directory can be queried. For example, the subdirectory color temp satisfies:

[0093] color_temp

[0094] —red_blue.jpg

[0095] —white_black.jpg

[0096] —pink.jpg

[0097] —text.txt

[0098] Among them, a text template list can be defined in the text.txt file, and the following text information can be listed in the text template list:

[0099] Label1: a red object with white background

[0100] Label2: a yellow object with white background

[0101] Label3: a blue object with white background

[0102] Label4: a green object with white background

[0103] In this way, four basic color attributes of red, yellow, blue, and green can be defined using the above text information as attribute recognition labels (Labels), and eigenvalue extraction can be performed on each of them respectively to generate four different template eigenvalues B1 - B4.

[0104] In addition, red_blue.jpg, as a picture file, can define an image that has both red and blue colors, and it can be separately extracted as a template eigenvalue B5. Among them, picture template files can be used to define attributes that cannot be effectively defined by text templates, thus greatly expanding the attribute matching range of templates. Specifically, as Figure 4 shown, taking the material as an example, a total of 8 images of materials including wood veneer, wooden floor, stainless steel, metal mesh, cloth texture, outdoor scene, marble, Changhong glass, and gradient glass paint are listed. Then, 8 jpg images need to be defined for each of the above materials in the directory of material temp, while the text list only needs to define simple and common materials.

[0105] Therefore, during the matching process, the template data of xxx.jpg and text.txt in all subdirectories can be counted separately. Assuming that there are 10, 5, and 3 templates for color, material, and shape attributes respectively, the target eigenvalues extracted from the target sub - graph to be recognized can be respectively calculated for similarity with the eigenvalues corresponding to these 10, 5, and 3 templates, and then the corresponding attribute recognition labels can be matched according to the calculation results of the similarity. Finally, according to the matched attribute recognition labels, the final matching result can be obtained for each attribute. For example, color is "red", material is "wooden", and shape is "square".

[0106] Therefore, with the help of the graphic - text matching strategy defined by the matching relationship between the target eigenvalue of the sub - graph to be recognized and the set of template eigenvalues of the preset matching template, parallel high - precision recognition of two or more object attribute features can be satisfied. For example, color, shape, texture, material, etc. can all be accurately recognized in parallel. In addition, only through some relatively general model tools, the corresponding attribute recognition can be achieved, thus ensuring the low - cost of the entire processing process while improving the intelligent level, and at the same time making it have strong scene versatility and a wider scope of application.

[0107] As Figures 2 - 4 shown, according to an embodiment of the present invention, in determining the attribute recognition label of the sub - graph to be recognized according to the similarity, it includes:

[0108] When the similarity is greater than or equal to the preset threshold, determine the attribute recognition label corresponding to the template eigenvalue of the similarity.

[0109] As described above, for the matching results of the similarity between the target eigenvalue corresponding to the sub-graph to be recognized and each template eigenvalue in the template eigenvalue set, a preset threshold based on the similarity can be used to extract the corresponding template eigenvalues in the matching results that are greater than or equal to the preset threshold, so as to obtain the attribute recognition labels of the corresponding templates according to these template eigenvalues.

[0110] For example, after performing similarity calculation and matching between the template eigenvalue Bx of 10 templates in color temp and the target eigenvalue A, it is found that the similarity values corresponding to the two colors "red" and "orange" in the corresponding attribute recognition labels exceed the preset threshold. Then, the attribute recognition labels corresponding to the two colors "red" and "orange", such as "a red object with white background" and "a orange object with white background", can be determined.

[0111] Therefore, the attribute recognition label that is closest to the target recognition attribute of the target recognition object can be accurately obtained.

[0112] As Figures 2 - 4 shown, according to an embodiment of the present invention, in operation S203 of determining the corresponding object attribute information according to the attribute recognition label, it includes:

[0113] Sorting the attribute recognition labels according to the similarity; determining the object attribute information corresponding to the sorted attribute recognition labels; or

[0114] Determining the object attribute information corresponding to the attribute recognition label with the maximum similarity.

[0115] The above matching strategy of the embodiment of the present invention can extract the corresponding template eigenvalues that are greater than or equal to the preset threshold in the similarity according to the similarity calculation results between the target eigenvalue and each template eigenvalue in the matching relationship. Among them, the template eigenvalue Bx can be understood as the feature similarity of the corresponding template, and the corresponding cosine similarity can be obtained by performing matching calculation between it and the corresponding target eigenvalue A. For the cosine similarities corresponding to B1 - Bx that exceed the similarity preset threshold, the labels (label1 - labelx) of the templates corresponding to the template eigenvalues B1 - Bx can be directly extracted as the attribute recognition labels, and each extracted attribute recognition label corresponds to a different cosine similarity. Therefore, according to the magnitudes of these cosine similarities, the above-mentioned attribute recognition labels can be sorted. Then, the corresponding object attribute information can be determined for these sorted attribute recognition labels in sequence.

[0116] Similarly, without considering the above similarity preset threshold, the matching results of all cosine similarities can be sorted for all calculated attribute recognition tags, and the corresponding object attribute information can be determined in sequence. Among them, the object attribute information can include the names, features, and even codes of all sorted object attributes, such as "color, red, red1 (code)". Further, the attribute recognition tag with the maximum similarity in the matching results can be directly queried, and its corresponding object attribute information can be output. For example, the target feature value of the sub-graph to be recognized can be calculated and matched with the n text information in the text template list of text.txt in the text template color_temp and the corresponding template feature values of multiple jpg in the image template respectively. Finally, an attribute recognition tag with the maximum cosine similarity can be selected as the color tag. Correspondingly, the recognition of material (material_temp) and shape (shape_temp) can also be implemented in parallel for matching, which will not be elaborated here. Therefore, different object attribute information can be pushed and displayed to the user in different ways, thus providing a more intelligent user experience.

[0117] Therefore, parallel matching and recognition of multiple attributes can be achieved simultaneously, effectively avoiding the situation in the prior art where independent data preparation and repetitive work are required for the recognition of each attribute, greatly improving the processing accuracy and efficiency of the image processing process, and also significantly improving the intelligence level of the intelligent agent.

[0118] As Figures 2 - 4 shown, according to an embodiment of the present invention, the image processing method applied to object attribute recognition further includes: updating the corresponding attribute recognition tag according to the dynamic detection result of the preset matching template.

[0119] During the operation of the intelligent agent, the file content in a specific preset matching template directory can be parsed when the system is started. As mentioned above, each sub-directory represents the recognition of a type of attribute, and each attribute will give a category with a corresponding similarity through the above matching strategy.

[0120] During the operation of the system, each sub-directory of the preset matching template can be monitored in real time. Once an operation such as addition, deletion, modification, or query of a folder or file in the corresponding sub-directory is detected (for example, when the user updates the file content or even the file structure of the preset matching template, corresponding dynamic detection results will be generated. The dynamic detection results are dynamic instructions generated by detecting the above-mentioned modification operations on the preset matching template). According to this dynamic instruction, an automatic loading mechanism similar to hot loading can be automatically triggered, and the system will reload the preset matching template after modification and update. For example, if a line "a black object with white background" is added to text.txt in the color directory color temp, the system can re-execute the loading process of the preset matching template through an automatic loading mechanism such as hot loading, so as to automatically complete the recognition of new attributes (color black). Specifically, the loading target can be the text or image template of the corresponding directory after modification, or the automatic recording of new attribute templates for new folders. For example, adding a matching template for "texture" on the basis of the matching template of "color, material, and shape" can also be achieved through the above loading mechanism.

[0121] Therefore, the above method of the embodiment of the present invention can use the automatic loading mechanism to perform online update of the corresponding template, so as to arbitrarily expand the categories of attributes to be recognized, including the new loading of text templates and image templates, and will not affect the processing process of the current process, thus achieving a low-cost operation effect.

[0122] In summary, the image processing method for object attribute recognition provided by the embodiment of the present invention can meet the parallel high-precision recognition of two or more object attribute features by means of the graphic matching strategy defined by the matching relationship between the target feature value of the sub-graph to be recognized and the set of template feature values of the preset matching template. For example, color, shape, texture, material, etc. can all be accurately recognized in parallel; at the same time, during the image processing process, background noise can be effectively reduced, greatly improving the accuracy of object attribute recognition; moreover, for the newly added attribute features in the preset matching template, recognition can also be directly achieved without performing repetitive work such as iterative tuning, and the attribute categories can be arbitrarily expanded, with stronger versatility; further, combined with the hot loading mechanism, online update of the newly added attribute categories can be realized at low cost. Therefore, the above image processing method of the embodiment of the present invention can meet the requirement of improving the recognition versatility of the entire system at low cost, with higher recognition accuracy, thus greatly improving the intelligent level of recognition.

[0123] Based on the above image processing method for object attribute recognition, the present invention also provides an image processing device for object attribute recognition. The following will be combined with Figure 5 This device will be described in detail.

[0124] Figure 5 A structural block diagram of an image processing apparatus applied to object attribute recognition according to an embodiment of the present invention is schematically shown.

[0125] As Figure 5 shown, the image processing apparatus 500 applied to object attribute recognition in this embodiment includes a sub - graph acquisition module 510, a label determination module 520, and an information determination module 530.

[0126] The sub - graph acquisition module 510 is used to acquire a sub - graph to be recognized of a target recognition image. In one embodiment, the sub - graph acquisition module 510 can be used to perform the operation S201 described above, which will not be elaborated here.

[0127] The label determination module 520 is used to determine an attribute recognition label of the sub - graph to be recognized according to the matching relationship between the target feature value of the sub - graph to be recognized and the set of template feature values of a preset matching template. In one embodiment, the label determination module 520 can be used to perform the operation S202 described above, which will not be elaborated here.

[0128] The information determination module 530 is used to determine corresponding object attribute information according to the attribute recognition label. In one embodiment, the information determination module 530 can be used to perform the operation S203 described above, which will not be elaborated here.

[0129] According to an embodiment of the present invention, any multiple of the sub - graph acquisition module 510, the label determination module 520, and the information determination module 530 can be combined and implemented in one module, or any one of them can be split into multiple modules. Or, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present invention, at least one of the sub - graph acquisition module 510, the label determination module 520, and the information determination module 530 can be at least partially implemented as a hardware circuit, such as a field - programmable gate array (FPGA), a programmable logic array (PLA), a system - on - chip, a system - on - substrate, a system - on - package, an application - specific integrated circuit (ASIC), or can be implemented by any other reasonable means of integrating or packaging circuits, etc., in hardware or firmware, or can be implemented in any one of the three implementation manners of software, hardware, and firmware, or in an appropriate combination of any several of them. Or, at least one of the sub - graph acquisition module 510, the label determination module 520, and the information determination module 530 can be at least partially implemented as a computer program module, and when the computer program module is run, it can execute the corresponding function.

[0130] Figure 6A block diagram of an electronic device suitable for implementing an image processing method applied to object attribute recognition according to an embodiment of the present invention is schematically shown.

[0131] The above-mentioned electronic device provided by the embodiment of the present invention includes one or more processors and a memory. The memory is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors are caused to execute the above-mentioned image processing method applied to object attribute recognition.

[0132] As Figure 6 shown, the electronic device 600 according to an embodiment of the present invention includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage section 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general microprocessor (such as a CPU), an instruction set processor, and / or a related chipset, and / or a dedicated microprocessor (such as an application specific integrated circuit (ASIC)), etc. The processor 601 may also include on-board memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0133] In the RAM 603, various programs and data required for the operation of the electronic device 600 are stored. The processor 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. The processor 601 executes various operations of the method flow according to an embodiment of the present invention by executing the programs in the ROM 602 and / or the RAM 603. It should be noted that the programs may also be stored in one or more memories other than the ROM 602 and the RAM 603. The processor 601 may also execute various operations of the method flow according to an embodiment of the present invention by executing the programs stored in the one or more memories.

[0134] According to an embodiment of the present invention, the electronic device 600 may further include an input / output (I / O) interface 605, and the input / output (I / O) interface 605 is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the I / O interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. The driver 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the driver 610 as needed, so that a computer program read from it can be installed into the storage portion 608 as needed.

[0135] The present invention also provides a computer-readable storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor is caused to execute the above-mentioned image processing method applied to object attribute recognition.

[0136] Among them, the computer-readable storage medium may be included in the device / apparatus / system described in the above embodiment; or it may exist alone and not be assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the one or more programs are executed, the method according to the embodiment of the present invention is implemented.

[0137] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer-readable storage medium may be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present invention, the computer-readable storage medium may include the above-described ROM 602 and / or RAM 603 and / or one or more memories other than ROM 602 and RAM 603.

[0138] An embodiment of the present invention further includes a computer program product, which includes a computer program, and when the computer program is executed by a processor, the above-mentioned image processing method applied to object attribute recognition is implemented.

[0139] Among them, the computer program includes program code for executing the method shown in the flowchart. When the computer program product runs in a computer system, the program code is used to enable the computer system to implement the method provided by the embodiments of the present invention.

[0140] When the computer program is executed by the processor 601, the above functions defined in the system / apparatus of the embodiments of the present invention are executed. According to the embodiments of the present invention, the above-described systems, apparatuses, modules, units, etc. can be implemented by computer program modules.

[0141] In one embodiment, the computer program can rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program can also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 609, and / or installed from the removable medium 611. The program code included in the computer program can be transmitted by any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0142] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, the above functions defined in the system of the embodiments of the present invention are executed. According to the embodiments of the present invention, the above-described systems, devices, apparatuses, modules, units, etc. can be implemented by computer program modules.

[0143] According to the embodiments of the present invention, the program code for executing the computer program provided by the embodiments of the present invention can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level procedures and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages include but are not limited to, such as Java, C++, python, the "C" language, or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, by using an Internet service provider to connect through the Internet).

[0144] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks may occur in a different order than noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, as well as combinations of blocks in the block diagram or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or by a combination of dedicated hardware and computer instructions.

[0145] Those skilled in the art will appreciate that the features recited in the various embodiments and / or claims of the present invention may be combined in various ways and / or combinations, even if such combinations or combinations are not explicitly recited in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features recited in the various embodiments and / or claims of the present invention may be combined in various ways and / or combinations. All such combinations and / or combinations fall within the scope of the present invention.

[0146] The embodiments of the present invention have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. Although the embodiments have been described separately above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of the present invention is defined by the appended claims and their equivalents. Without departing from the scope of the present invention, those skilled in the art can make various substitutions and modifications, and all such substitutions and modifications should fall within the scope of the present invention.

Claims

1. An image processing method applied to object attribute recognition, characterized in that, Comprising: Obtaining a sub - graph to be recognized with a solid - color background corresponding to the target recognition image; Determining an attribute recognition label of the sub - graph to be recognized according to the matching relationship between the target feature value of the sub - graph to be recognized and the set of template feature values of a preset matching template, including: obtaining the similarity between the target feature value and each template feature value in the set of template feature values; determining the attribute recognition label of the sub - graph to be recognized according to the similarity, where the attribute recognition label is preset text or picture information directly related to the attribute in the preset matching template and corresponds to the feature value of the corresponding attribute; and Determining corresponding object attribute information according to the attribute recognition label; Updating the corresponding attribute recognition label according to the dynamic detection result of the preset matching template; During the operation of the intelligent agent, when the system starts, it parses the file content in the preset matching template directory; Each sub - directory represents the recognition of a class of attributes, and each attribute will be matched through the above - mentioned text - image matching strategy to give a category with a corresponding similarity; The dynamic detection result is a dynamic instruction generated by detecting the modification operation of the preset matching template. According to this dynamic instruction, an automatic loading mechanism for hot - loading is automatically triggered, and the system reloads the preset matching template after modification and update; Wherein, the target feature value is the feature value of the sub - graph to be recognized related to the target recognition attribute corresponding to the target recognition object, the preset matching template is template data with corresponding attribute recognition labels, these template data are used to define the corresponding feature values of each attribute feature, and the set of template feature values includes all feature values of the text and picture templates in the preset matching template; Wherein, each template feature value in the set of template feature values of the preset matching template is also a value corresponding to the same dimension as the target feature value; The target feature value and each template feature value are matched, and the cosine similarity is calculated between the target feature value and each template feature value in the set of template feature values respectively, so as to determine the corresponding similarity.

2. The method according to claim 1, wherein In the obtaining of the sub - graph to be recognized of the target recognition image, it includes: Obtaining the mask data information of the target recognition image; Obtaining the sub - graph to be recognized of the target recognition image according to the mask data information.

3. The method according to claim 2, characterized in that, In the obtaining of the mask data information of the target recognition image, it includes: Obtaining the category sub - graph of the target recognition image that matches the preset category information; Extracting the mask data information of the category sub - graph.

4. The method according to claim 2, wherein In the obtaining of the sub - graph to be recognized of the target recognition image according to the mask data information, it includes: Generating a background image corresponding to the mask data information; Obtaining the sub - graph to be recognized according to the background image.

5. The method according to claim 1, characterized in that In the determining of the attribute recognition label of the sub - graph to be recognized according to the similarity, it includes: When the similarity is greater than or equal to a preset threshold, determining the attribute recognition label of the template feature value corresponding to the similarity.

6. The method according to claim 5, wherein In the determining of the corresponding object attribute information according to the attribute recognition label, it includes: sorting the attribute recognition labels according to the similarity; determining the object attribute information corresponding to the sorted attribute recognition labels; or Determining the object attribute information corresponding to the attribute recognition label with the maximum similarity.

7. An image processing apparatus for object attribute recognition, which is used to implement the method described in claim 1, characterized in that, Comprising: Sub - figure acquisition module, configured to acquire a sub - figure to be recognized of a target recognition image; Label determination module, configured to determine an attribute recognition label of the sub - figure to be recognized according to the matching relationship between the target feature value of the sub - figure to be recognized and the set of template feature values of a preset matching template, including: obtaining the similarity between the target feature value and each template feature value in the set of template feature values; determining the attribute recognition label of the sub - figure to be recognized according to the similarity; and Information determination module, configured to determine corresponding object attribute information according to the attribute recognition label; Wherein, the target feature value is a feature value of the sub - figure to be recognized related to the target recognition attribute corresponding to the target recognition object, the preset matching template is template data with corresponding attribute recognition labels, these template data are used to define the corresponding feature values of each attribute feature, and the set of template feature values includes all feature values of the text and picture templates in the preset matching template; Wherein, each template feature value in the set of template feature values of the preset matching template is also a value corresponding to the dimension of the target feature value; the target feature value and each template feature value are matched, and the cosine similarity is calculated by the target feature value and each template feature value in the set of template feature values respectively, so as to determine the corresponding similarity.

8. An electronic device, comprising: One or more processors; A memory for storing one or more programs, Wherein, when the one or more programs are executed by the one or more processors, the one or more processors execute the method according to any one of claims 1 - 6.

9. A computer - readable storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor executes the method according to any one of claims 1 - 6.

10. A computer program product, comprising a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 - 6 is implemented.

Images