Image processing method, device, equipment, and storage medium

Abstract

The present disclosure provides an image processing method, device, equipment, and storage medium, the method including: determining a target image whose image resolution meets a preset resolution condition based on the shooting screen of the shooting page when a preset recognition operation is received triggered for the shooting screen of the shooting page, and then sending the target image to an image recognition service side to recognize the target image. As described above, the image processing method according to the embodiment of the present disclosure first determines a target image whose image resolution meets a preset resolution condition before performing image recognition, and allows the image recognition service side to recognize based on the target image whose resolution meets the condition, thereby improving the accuracy of image recognition.

Description

[Technical Field]

[0001] This application claims priority to a Chinese patent application filed on July 12, 2021, bearing application number 202110783343.X and entitled "Image Processing Method, Apparatus, Device and Storage Medium," the entire contents of which are incorporated herein by reference.

[0002] The present disclosure relates to the field of data processing, and in particular to an image processing method, apparatus, device and storage medium. [Background technology]

[0003] As multimedia short videos become more popular, interest in taking pictures or videos is increasing, and the functions of the photo page are also becoming more and more. For example, how to improve the accuracy of image recognition for the recognition function of the photo page's photo screen is currently a technical issue that needs to be solved urgently. Summary of the Invention [Means for solving the problem]

[0004] In order to solve the above technical problems or at least partially solve the above technical problems, an embodiment of the present disclosure provides an image processing method, which allows the image recognition service side to recognize based on a target image whose image resolution meets a preset resolution condition, thereby improving the accuracy of image recognition.

[0005] In a first aspect, the present disclosure provides an image processing method, said method comprising: When receiving a preset recognition operation triggered for a capture screen of a capture page, determining a target image whose image resolution satisfies a preset resolution condition based on the capture screen of the capture page; and transmitting the target image to an image recognition service for recognizing the target image.

[0006] In an alternative embodiment, prior to the step of transmitting the target image to an image recognition service, extracting image features of the target image; matching the image features with feature parameters of pre-stored preset objects; If it is determined that the matching is successful, the method further includes determining a preset object corresponding to the successfully matched feature parameters as an initial recognition result of the target image.

[0007] In response to this, the step of transmitting the target image to an image recognition service side includes: The method includes transmitting the target image and an initial recognition result of the target image to an image recognition service side, which recognizes the target image based on the initial recognition result.

[0008] In an alternative embodiment, the step of transmitting the target image to an image recognition service comprises: The method further includes a step of transmitting the target image to the image recognition service side when it is determined that the image features have not matched with the feature parameters of the pre-stored preset object.

[0009] In an optional embodiment, when a preset recognition operation triggered for a capture screen of a capture page is received, the step of determining a target image whose image resolution satisfies a preset resolution condition based on the capture screen of the capture page includes: When receiving a preset recognition operation triggered for a photographing screen of a photographing page, obtaining a current image corresponding to the photographing screen, and determining whether the image resolution of the current image satisfies a preset resolution condition; If it is determined that the image resolution of the current image satisfies the preset resolution condition, the current image is determined to be a target image; otherwise, a current image corresponding to the photographed screen is acquired until it is determined that the number of times of acquiring a current preset image is reached, and the step of determining whether the image resolution of the current image satisfies the preset resolution condition is re-executed.

[0010] In an optional embodiment, when it is determined that the preset number of image acquisitions has been reached and an image that satisfies the preset resolution condition has not been determined, recognition failure hint information' is displayed on the photographing page.

[0011] In an optional embodiment, when a preset recognition operation triggered for a capture screen of a capture page is received, the step of determining a target image whose image resolution satisfies a preset resolution condition based on the capture screen of the capture page includes: When receiving a preset recognition operation triggered for a photographing screen of a photographing page, obtaining an image set to be processed including consecutive multi-frame images with a current image corresponding to the photographing screen as an end frame; and determining, as a target image, an image whose image resolution satisfies a preset resolution condition from the set of images to be processed.

[0012] In an optional embodiment, if no image that satisfies the preset resolution condition is determined from the set of images to be processed, a recognition failure hint information' is displayed on the capture page.

[0013] In a second aspect, the present disclosure further provides an image processing device, the device comprising: a determination module for determining, when receiving a preset recognition operation triggered for a capture screen of a capture page, a target image whose image resolution satisfies a preset resolution condition based on the capture screen of the capture page; and a transmitting module for transmitting the target image to an image recognition service for recognizing the target image.

[0014] In a third aspect, the present disclosure provides a computer-readable storage medium having instructions stored thereon that, when executed on a terminal device, cause the terminal device to implement the above method.

[0015] In a fourth aspect, the present disclosure provides an apparatus, comprising: a memory; a processor; and a computer program stored in the memory and executable by the processor, the computer program, when executed by the processor, realizing the method described above.

[0016] In a fifth aspect, the present disclosure provides a computer program product, said computer program product comprising computer programs / instructions which when executed by a processor implement the above method.

[0017] The technical solutions according to the embodiments of the present disclosure have the following advantages over the prior art:

[0018] An embodiment of the present disclosure provides an image processing method, which first determines a target image whose image resolution meets a preset resolution condition based on the captured image of the captured page when a preset recognition operation is received for the captured image of the captured page, and then sends the target image to an image recognition service side for recognition. Thus, the image processing method according to the embodiment of the present disclosure first determines a target image whose image resolution meets a preset resolution condition before performing image recognition, and allows the image recognition service side to perform recognition based on the target image whose resolution meets the condition, thereby improving the accuracy of image recognition.

[0019] Furthermore, since the recognized target image satisfies the preset resolution condition, the occurrence of invalid inputs and invalid recognitions is reduced, and the overall efficiency of image recognition is improved.

[0020] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure.

[0021] In order to more clearly describe the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly describes the drawings that need to be used to describe the embodiments or the prior art, and obviously, a person skilled in the art can also obtain other drawings based on these drawings without any creative efforts. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a flowchart of an image processing method according to an embodiment of the present disclosure. [Figure 2] 10 is a flowchart of another image processing method according to an embodiment of the present disclosure. [Figure 3] 10 is a flowchart of another image processing method according to an embodiment of the present disclosure. [Figure 4] 10 is a flowchart of another image processing method according to an embodiment of the present disclosure. [Figure 5] 1 is a structural schematic diagram of an image processing device according to an embodiment of the present disclosure. [Figure 6] 1 is a structural schematic diagram of an image processing device according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0023] In order to make the above-mentioned objects, features and advantages of the present disclosure more clearly understood, the solutions of the present disclosure are further described below. As needed to explain, the embodiments of the present disclosure and the features in the embodiments can be combined with each other when not contradictory.

[0024] In the following description, many specific details are set forth in order to fully understand the present disclosure, but the present disclosure may be implemented in ways different from those described herein. Obviously, the embodiments in the specification are only a part of the embodiments of the present disclosure, but not all of the embodiments.

[0025] To improve the accuracy of image recognition, an embodiment of the present disclosure proposes an image processing method that first determines a target image whose image resolution meets a preset resolution condition based on the captured image of the captured page when a preset recognition operation is received for the captured image of the captured page. Then, the target image is sent to an image recognition service side for recognition. As described above, the image processing method according to the embodiment of the present disclosure first determines a target image whose image resolution meets a preset resolution condition before performing image recognition, and allows the image recognition service side to perform recognition based on the target image whose resolution meets the condition, thereby improving the accuracy of image recognition.

[0026] Furthermore, since the recognized target image satisfies the preset resolution condition, the occurrence of invalid inputs and invalid recognitions is reduced, and the overall efficiency of image recognition is improved.

[0027] On this basis, an embodiment of the present disclosure provides an image processing method, and FIG. 1 is a flowchart of the image processing method according to the embodiment of the present disclosure, which includes the following steps S101 to S102.

[0028] S101, when receiving a preset recognition operation triggered for a photographed image of a photographed page, determine a target image whose image resolution satisfies a preset resolution condition based on the photographed image of the photographed page.

[0029] In the embodiments of the present disclosure, the method of triggering a preset recognition operation on the shooting screen of the shooting page can include multiple methods, for example, method 1 is to trigger a long press operation on the shooting screen of the shooting page, and the long press operation can be triggered on any position on the shooting screen; method 2 is to trigger an operation on a recognition control installed on the shooting page, and the recognition control is installed at any position on the shooting page (for example, a position on the right side, bottom, etc. of the shooting page), etc.

[0030] In the embodiment of the present disclosure, when a preset recognition operation is triggered for the photographic image of the photographing page, an image whose image resolution meets the preset resolution requirement is determined as the target image based on the photographic image of the photographing page. The specific implementation method will be described in the subsequent embodiment, and detailed description thereof will be omitted here.

[0031] The capture screen of the capture page may include one frame or a series of multi-frame images captured with the current camera.

[0032] In an embodiment of the present disclosure, an image whose image resolution satisfies a preset resolution condition is determined as a target image based on the captured image of the capture page. In a subsequent recognition process, recognition is performed based on the target image that satisfies the preset resolution condition, which contributes to reducing the occurrence of invalid inputs. The method for determining whether the image resolution of an image satisfies the preset resolution condition may include processing the image based on an image resolution model to obtain a target image that satisfies the preset resolution condition.

[0033] For example, assuming that an image resolution model is deployed on a client (such as a mobile phone, tablet computer, or PC) that has a photographing function (such as a camera), the image resolution model is trained based on image samples with different resolutions, and after a target image is processed by the image resolution model, a score for the target image can be output. If the threshold for the image resolution score is set to 80 points, an image that meets the preset resolution conditions refers to an image whose corresponding image resolution score is 80 points or higher.

[0034] As should be explained, the preset image resolution score threshold is used to determine whether the image resolution of an image meets the preset resolution conditions, and can be set to 80 points, 85 points, 90 points, etc. as needed, and the embodiments of the present disclosure do not limit the preset image resolution score threshold.

[0035] S102: The target image is transmitted to the image recognition service side.

[0036] The image recognition service is used to recognize the target image.

[0037] In an embodiment of the present disclosure, the target image determined in S101 above is sent to an image recognition service side, which then recognizes the target image. The image recognition service side is communicatively connected to a client, and based on the communication connection between the image recognition service side and the client, the client can send the target image to the image recognition service side, and the image recognition service side can return the recognition result and / or a recommended template based on the recognition result to the client.

[0038] In an embodiment of the present disclosure, if the target image is successfully recognized, the image recognition service may return the recognition result and / or a recommended template based on the recognition result to the client. For example, the image recognition service may return to the client explanatory information about the recognition result "Sunflower" corresponding to the target image, or a recommended template corresponding to the recognition result "Sunflower" corresponding to the target image. If the target image is not successfully recognized, the image recognition service may return a recognition failure command to the client, prompting the user to fail the current image recognition attempt, and may return a recommended generic template recognized for the generic scene to the client. For example, the image recognition service may prompt the user by flashing text on the capture page or in the form of a Toast pop-up window. Specifically, the prompt text may be "The matching result was not recognized. We recommend a generic template or please try recognition again!", and a generic template may be displayed at the bottom of the capture page.

[0039] In an image processing method according to an embodiment of the present disclosure, when a preset recognition operation is received for a captured image of a captured page, a target image whose image resolution satisfies a preset resolution condition is first determined based on the captured image of the captured page. The target image is then sent to an image recognition service for recognition. As described above, the image processing method according to an embodiment of the present disclosure first determines a target image whose image resolution satisfies a preset resolution condition before performing image recognition, allowing the image recognition service to recognize the target image based on the target image whose resolution satisfies the condition, thereby improving the accuracy of image recognition. Furthermore, because the recognized target image satisfies the preset resolution condition, the occurrence of invalid inputs and invalid recognition is reduced, improving the overall efficiency of image recognition.

[0040] Based on the above embodiment, the method for determining whether the image resolution satisfies the preset resolution condition may include a number of implementation methods, and two specific implementation methods are listed below.

[0041] In an alternative embodiment, FIG. 2 is a flowchart of another image processing method according to an embodiment of the present disclosure, in which the above S101 specifically includes the following S201 to S203.

[0042] S201, when a preset recognition operation triggered on a photographing screen of a photographing page is received, an image set to be processed is obtained.

[0043] The image set to be processed includes a series of multi-frame images, the end frame of which is the current image corresponding to the photographed scene.

[0044] In an embodiment of the present disclosure, when a preset recognition operation triggered for the shooting screen of the shooting page is received, an image set to be processed is obtained, and the image set to be processed includes consecutive multi-frame images with the current image corresponding to the shooting screen as the end frame. Typically, the preview images in the camera preview interface are stored in the form of a preview stream, and when a preset recognition operation triggered for the shooting screen of the shooting page is received, the latest N frames of preview images can be obtained from the stored preview pictures in the preview stream format to form an image set to be processed together with the current image.

[0045] As should be explained, the consecutive multi-frame images included in the image set to be processed, with the current image corresponding to the captured screen as the end frame, may be consecutive 5-frame images, consecutive 7-frame images, etc., and the embodiments of the present disclosure do not limit the specific number of frames in the consecutive multi-frame images.

[0046] S202: Determine whether or not there is an image in the image set to be processed whose image resolution satisfies the preset resolution conditions. If it is determined that there is an image (YES), use the image that satisfies the preset resolution conditions as the target image and execute S102. If it is not determined that there is an image (NO), execute S203.

[0047] In an embodiment of the present disclosure, based on an image recognition model, each image in the image set to be processed can be processed to determine whether there is an image in the image set to be processed whose image resolution meets a preset resolution condition.

[0048] For example, an image resolution model is placed on the client, and the threshold value of the image resolution score is set to 80 points. Images that satisfy the preset resolution conditions are those whose corresponding image resolution scores are 80 points or more. Specifically, the image set to be processed (assumed to include images 1 to 5) obtained in S201 is input into the image resolution model, and an image resolution score for each image is obtained based on the processing of the image resolution model. Each image resolution score is compared with a preset threshold value (e.g., 80 points), and it is determined whether the image set to be processed includes any images that satisfy the preset resolution conditions.

[0049] For example, if the image resolution score of only one frame image in the image set to be processed is equal to or greater than the image resolution threshold, i.e., if only one frame image satisfies the preset resolution condition, then that image is directly determined as the target image. Suppose the image resolution scores of images 1 to 5 are 70, 74, 76, 79, and 82, respectively, then the image resolution score of image 5 is greater than the preset threshold (e.g., 80 points), i.e., image 5 is an image that satisfies the preset resolution condition. After determining image 5 as the target image, S102 in the above embodiment continues to be executed.

[0050] If the image resolution scores of the multi-frame images in the image set to be processed are equal to or greater than the image resolution threshold, i.e., if the multi-frame images in the image set to be processed satisfy the preset resolution conditions, the corresponding images with higher image resolution scores can be used as the target images. Suppose the image resolution scores of images 1 to 5 are 70, 74, 78, 82, and 90, respectively. Since the image resolution scores of images 4 and 5 are all greater than the preset threshold (e.g., 80 points), i.e., images 4 and 5 satisfy the preset resolution conditions and have a high image resolution score, image 5 can be selected as the target image. After image 5 is determined as the target image, step S102 in the above embodiment continues to be executed.

[0051] Alternatively, one image can be randomly determined as the target image from a number of images in the image set to be processed that meet a preset resolution condition.

[0052] If all of the image resolution scores corresponding to the images in the image set to be processed are smaller than the image resolution threshold, that is, if no images satisfying the preset resolution conditions have been determined in the image set to be processed, S203 is executed.

[0053] S203, display 'recognition failure hint information' on the photographing page.

[0054] In the embodiment of the present disclosure, if an image that satisfies the preset resolution condition is not determined from the image set to be processed, a recognition failure hint information' is displayed on the capture page. For example, the hint information 'Recognition failed, please re-trigger the recognition function' is displayed on the capture page and flashes at a preset interval time, or a Toast pop-up window is popped up on the capture page to prompt the user to fail the recognition.

[0055] As described above, in an embodiment of the present disclosure, when the client has not determined a target image whose resolution meets the preset image resolution conditions, the client does not send the target image to the image recognition service side, thereby reducing the number of invalid recognitions on the image recognition service side and reducing waste of network resources between the client and the image recognition service side.

[0056] In the above-mentioned optional embodiment, when a preset recognition operation triggered for the shooting screen of the shooting page is received, a set of images to be processed including consecutive multi-frame images is obtained and used to determine the target image. However, since all of the set of images to be processed are stored in the client, there is a possibility that the memory usage will be too large, resulting in a slowdown in the execution speed of the client system. Therefore, in order to reduce the load on the client's memory, when a preset recognition operation triggered for the shooting screen of the shooting page is received, only one frame of the current image corresponding to the shooting screen can be obtained each time.

[0057] For this reason, the embodiment of the present disclosure provides a flowchart of another image processing method, which, referring to FIG. 3, includes the following steps S301 to S304.

[0058] S301, when receiving a preset recognition operation triggered for a shooting screen of a shooting page, obtain a current image corresponding to the shooting screen.

[0059] In an embodiment of the present disclosure, when a preset recognition operation triggered for a shooting screen of a shooting page is received, a current image corresponding to the shooting screen is obtained, and the current image corresponding to the shooting screen includes a one-frame image, such as image 1.

[0060] S302: Determine whether the image resolution of the current image satisfies the preset resolution conditions. If the conditions are met (YES), determine the current image as the target image and execute S102. If the conditions are not met (NO), execute S303.

[0061] In an embodiment of the present disclosure, an acquired current image can be processed based on an image recognition model to determine whether the image resolution of the current image satisfies a preset resolution condition. For example, an image resolution model is configured on a client, and the image resolution score threshold is set to 80 points. An image that satisfies the preset resolution condition is an image whose corresponding image resolution score is 80 points or greater. Specifically, the current image of one frame acquired in S301 is input into the image resolution model, and the image resolution score of the current image of the frame is obtained based on the processing of the image resolution model. The image resolution score is compared with a preset threshold (e.g., 80 points) to determine whether the image resolution of the current image of the frame satisfies the preset resolution condition.

[0062] If the image resolution score of the current image is equal to or greater than the image resolution threshold, the current image is determined as the target image, and then S102 in the above embodiment is continued to be executed; otherwise, it is determined that the current image does not meet the preset resolution condition, and S303 is executed.

[0063] In S303, it is determined whether the number of times the current image has been acquired has reached the preset number of times the image has been acquired. If NO, S301 is executed; if YES, S304 is executed.

[0064] In the embodiment of the present disclosure, after it is determined that the image resolution of the current image does not satisfy the preset resolution condition, it is determined whether the number of times the current image is captured reaches the preset number of times of image capture, i.e., the number of times the current image is captured is compared with the preset number of times of image capture, and the preset number of times of image capture may be 3, 5, 7, etc., and the embodiment of the present disclosure does not limit the preset number of times of image capture.

[0065] For example, assuming that the preset image acquisition count is three, when a preset recognition operation triggered for the shooting screen of the shooting page is received, a current image (e.g., image 1) corresponding to the shooting screen is acquired for the first time, and if the image resolution score of the obtained image 1 is smaller than the preset threshold (e.g., 80 points) based on the processing of the image resolution model, it is determined that the image resolution of image 1 does not satisfy the preset resolution condition. Furthermore, if it is determined that the number of times the current image is acquired (a total of 1 time) does not reach the preset image acquisition count (e.g., 3 times), S301 continues to be executed, that is, a current image (e.g., image 2) corresponding to the shooting screen is acquired for the second time, and if the image resolution score of the obtained image 2 is smaller than the preset threshold (e.g., 80 points) based on the processing of the image resolution model, it is determined that the image resolution of image 2 does not satisfy the preset resolution condition. Furthermore, if it is determined that the number of times the current image is acquired (a total of two times) does not reach the preset number of times of image acquisition (e.g., three times), S301 continues to be executed, and the current image (e.g., image 3) corresponding to the captured screen is acquired for the third time, and if the image resolution score of the acquired image 3 is smaller than the preset threshold (e.g., 80 points) based on the processing of the image resolution model, it is determined that the image resolution of image 2 does not satisfy the preset resolution condition. Furthermore, if it is determined that the number of times the current image is acquired (a total of three times) reaches the preset number of times of image acquisition (e.g., three times), there is no need to re-acquire the current image corresponding to the captured screen, and S304 is executed to prompt the user to fail this recognition.

[0066] S304: Display 'recognition failure hint information' on the photographing page.

[0067] In the embodiment of the present disclosure, when it is determined that the number of times of current image capture reaches the preset number of image captures and an image that meets the preset resolution condition has not been determined, a recognition failure hint information' is displayed on the photographing page. The specific prompt method is shown in the example of S203 above, and a detailed description thereof is omitted here.

[0068] In an image processing method according to an embodiment of the present disclosure, when a preset recognition operation is triggered for a photographing screen on a photographing page, a current image corresponding to the photographing screen is acquired, and it is determined whether the image resolution of the current image satisfies a preset resolution condition. If the condition is met, the current image is determined as a target image. If the condition is not met, a current image corresponding to the photographing screen is acquired, and this process continues until the image resolution of the acquired current image satisfies the preset resolution condition or the number of acquisitions of the acquired current image reaches the preset number of image acquisitions. Furthermore, the target image is sent to the image recognition service side, or recognition failure hint information' is displayed on the photographing page. As described above, the image processing method according to an embodiment of the present disclosure can alleviate the problem of excessive memory usage when determining image resolution, resulting in a decrease in the execution speed of the client system.

[0069] In order to further improve the recognition efficiency on the image recognition service side, the target image is sent to the image recognition service side, and before image recognition is performed, the client first performs initial recognition on the target image, and then the initial recognition result is sent to the image recognition service side along with the target image, and the image recognition service side combines the target image and the initial recognition result to perform image recognition, thereby improving the recognition efficiency of the target image.

[0070] In an optional embodiment, a method for performing initial recognition on a target image may include first extracting image features of the target image, then matching the image features with feature parameters of a pre-stored preset object, and if it is determined that the matching is successful, determining the preset object corresponding to the successfully matched feature parameters as an initial recognition result for the target image, and further transmitting the target image and the initial recognition result for the target image to an image recognition service side, and recognizing the target image based on the initial recognition result.

[0071] In the embodiment of the present disclosure, after extracting image features of a target image, the extracted image features are matched with feature parameters of pre-stored preset objects to perform initial recognition of the target image. The feature parameters of the preset objects may include correspondences between the preset objects and the feature parameters. For example, the preset objects may include "flowers," "grass," "cats," etc., and the feature parameters of the preset objects may include correspondences between "flowers" and the feature parameters of "flowers," between "grass" and the feature parameters of "grass," between "cats" and the feature parameters of "cats," etc.

[0072] In an embodiment of the present disclosure, when it is determined that the image features of the target image are successfully matched with the feature parameters of a pre-stored preset object, the preset object corresponding to the successfully matched feature parameters is determined as the initial recognition result of the target image. As can be understood, the more pre-stored feature parameters corresponding to a preset object, the higher the accuracy of the initial recognition result of the preset object.

[0073] In an alternative embodiment, performing initial recognition on the image may be accomplished based on an image recognition model, for example, deployed on the client and trained based on image samples with preset objects.

[0074] For example, an image recognition model deployed on a client is used to perform initial recognition of a target image, and the extracted image features of the target image are characterized by the target image having the characteristics of a "flower" through processing of the image recognition model. If it is determined that the image features have been successfully matched with the pre-stored feature parameters of a "flower," the preset object "flower" is determined as the initial recognition result of the target image.

[0075] In an embodiment of the present disclosure, after determining an initial recognition result corresponding to a target image, the target image and the initial recognition result corresponding to the target image are sent to an image recognition service side, and the image recognition service side recognizes the target image based on the initial recognition result. For example, if the initial recognition result corresponding to the target image is determined to be a "flower," a type identifier of the "flower" is sent to the image recognition service side along with the target image, and the image recognition service side may include image recognition models for recognizing different objects (such as a "flower recognition" model, a "grass recognition" model, and a "cat recognition" model). Furthermore, the "flower recognition" model on the image recognition service side is invoked based on the received type identifier of the "flower" to recognize the target image and obtain a final recognition result. For example, the recognition result corresponding to the target image may be a "sunflower."

[0076] In an embodiment of the present disclosure, before a target image is sent to an image recognition service for image recognition, the client first performs initial recognition based on the image features of the target image to obtain an initial recognition result, and then sends the initial result and the target image to the image recognition service. Because the initial recognition result is obtained based on the image features of the target image and can directly characterize the features of the target image, the image recognition service can recognize the target image based on the initial recognition result, thereby improving the accuracy of image recognition. Furthermore, the image recognition service can call up a corresponding recognition model based on the initial recognition result and further intentionally recognize the target image, thereby improving the overall efficiency of image recognition.

[0077] In another optional embodiment, during the initial recognition process of the target image, if it is determined that the image features of the target image have failed to match the feature parameters of the pre-stored preset object, the target image can be sent individually to the image recognition service side, and the image recognition service side can recognize the target image based only on the target image.

[0078] In an embodiment of the present disclosure, the image recognition service side may further include a general-purpose recognition model for recognizing an object that fails to match with the feature parameters of the preset object. Specifically, if it is determined that the result of matching the image features with the feature parameters of the pre-stored preset object is a match failure, the target image is individually sent to the image recognition service side, and the image recognition service side recognizes the target image based on the general-purpose recognition model. After the recognition is successful, the recognition result or a template recommended based on the recognition result is returned to the client.

[0079] Based on the above embodiment, an embodiment of the present disclosure further provides an image processing method, taking the example of simultaneously configuring an image resolution model and an image recognition model in a client. FIG. 4 is a flowchart of another image processing method according to an embodiment of the present disclosure, which includes the following steps S401 to S408:

[0080] S401, when receiving a preset recognition operation triggered for a shooting screen of a shooting page, obtain a current image corresponding to the shooting screen.

[0081] S402: Determine whether the image resolution of the current image satisfies the preset resolution conditions. If the conditions are met (YES), determine the current image as the target image and execute S405. If the conditions are not met (NO), execute S403.

[0082] In S403, it is determined whether the number of times the current image has been acquired has reached the preset number of times the image has been acquired. If NO, S401 is executed; if YES, S404 is executed.

[0083] S404: Display recognition failure hint information on the photographing page.

[0084] The specific steps of S401 to S404 are described in detail in the above embodiment S301 to S304, and detailed description thereof will be omitted here.

[0085] S405, extracting image features of the target image.

[0086] S406: Match the image features with the feature parameters of the pre-stored preset object, and determine whether the matching result is successful. If yes, execute S407; if no, execute S408.

[0087] In an embodiment of the present disclosure, for example, the feature parameters of the preset objects include a correspondence between "flower" and the feature parameters of "flower," a correspondence between "grass" and the feature parameters of "grass," etc. The image features are matched with the feature parameters of the pre-stored preset objects by processing the image recognition model arranged in the client, and if it is determined that the matching result between the image features of image 1 and the feature parameters of the pre-stored preset objects is successful, S407 is executed. If it is determined that the matching result between the image features of image 2 and the feature parameters of the pre-stored preset objects is unsuccessful, S408 is executed.

[0088] S407: Determine the preset object corresponding to the successfully matched feature parameters as the initial recognition result of the target image, send the target image and the initial recognition result of the target image to the image recognition service side, and recognize the target image based on the initial recognition result.

[0089] In an embodiment of the present disclosure, for example, if the preset object corresponding to the feature parameter of "flower" is "flower," the initial recognition result of image 1 is "flower," and the type identifier of "flower" is sent to the image recognition service side along with image 1. Image 1 is recognized based on the type identifier of "flower," and the final recognition result obtained is "sunflower." The image recognition service side returns explanatory information about "sunflower" or recommended templates related to the recognition result "sunflower" to the client.

[0090] S408: The target image is sent to the image recognition service side, and image recognition is performed.

[0091] In an embodiment of the present disclosure, for example, Image 2 is sent to the image recognition service side, and the image recognition service side recognizes Image 2 based on the generic recognition model, and then returns a recognition failure command to the client, and returns the recognition result or a recommended template based on the recognition result, etc. Specifically, a Toast pop-up window is displayed on the shooting page saying "The matching result is not recognized, we recommend a generic template or please try recognition again!", and the generic template is displayed at the bottom of the shooting page.

[0092] In an image processing method according to an embodiment of the present disclosure, when a preset recognition operation is triggered for a shooting screen of a shooting page, a current image corresponding to the shooting screen is acquired, and it is determined whether the image resolution of the current image satisfies a preset resolution condition. If the condition is met, the current image is determined as a target image. If the condition is not met, the current image corresponding to the shooting screen is acquired, and this process continues until the image resolution of the acquired current image satisfies the preset resolution condition or the number of acquisitions of the acquired current image reaches the preset image acquisition number. After the target image is determined, image features of the target image are extracted, and the image features are matched with feature parameters of a pre-stored preset object. The preset object corresponding to the successfully matched feature parameters is determined as an initial recognition result of the target image. The target image and the initial recognition result of the target image are then sent to the image recognition service side, and the target image is recognized based on the initial recognition result.

[0093] As described above, an embodiment of the present disclosure first determines a target image whose image resolution satisfies a preset resolution condition before performing image recognition, performs initial recognition on the target image based on the image features of the target image, and then has the image recognition service perform image recognition on the target image based on the initial recognition results, thereby improving the accuracy of image recognition.

[0094] In addition, since the recognized target image satisfies the preset resolution requirements, the occurrence of invalid input and invalid recognition is reduced, and the overall efficiency of image recognition is improved. In addition, in addition to improving the accuracy of image recognition, the number of recognition failure prompts that users receive during the image recognition operation is reduced, thereby improving the user experience.

[0095] Based on the above method embodiment, the present disclosure further provides an image processing device. FIG. 5 is a structural schematic diagram of an image processing device according to an embodiment of the present disclosure, the device comprising: a first determination module 501 for determining, when receiving a preset recognition operation triggered for a photographing screen of a photographing page, a target image whose image resolution meets a preset resolution condition according to the photographing screen of the photographing page; and a sending module 502 for sending the target image to an image recognition service for recognizing the target image.

[0096] In an alternative embodiment, the apparatus comprises: an extraction module for extracting image features of the target image; a matching module for matching the image features with feature parameters of pre-stored preset objects; The apparatus further includes a second determination module for, when it is determined that the matching is successful, determining a preset object corresponding to the successfully matched feature parameters as an initial recognition result of the target image.

[0097] In response, the transmitting module 502: The first transmitting sub-module is configured to transmit the target image and an initial recognition result of the target image to an image recognition service side for recognizing the target image based on the initial recognition result.

[0098] In an alternative embodiment, the transmitting module 502: The image recognition service further includes a second transmission sub-module for transmitting the target image to the image recognition service side when it is determined that the image features have not matched with the feature parameters of the pre-stored preset object.

[0099] In an alternative embodiment, the first determination module 501: a first acquisition submodule for, when receiving a preset recognition operation triggered for a shooting screen of a shooting page, acquiring a current image corresponding to the shooting screen, and determining whether an image resolution of the current image satisfies a preset resolution condition; and a second determination sub-module for determining the current image as a target image if it is determined that the image resolution of the current image satisfies the preset resolution condition; otherwise, acquiring a current image corresponding to the shooting screen until it is determined that the current preset image acquisition number has been reached, and re-executing the step of determining whether the image resolution of the current image satisfies the preset resolution condition.

[0100] In an alternative embodiment, the apparatus comprises: The camera further includes a first prompt module for displaying 'recognition failure hint information' on the photographing page when it is determined that the preset number of image acquisitions has been reached and an image satisfying the preset resolution condition has not been determined.

[0101] In an alternative embodiment, the first determination module 501: a second acquisition sub-module for acquiring an image set to be processed, the image set including consecutive multi-frame images, with a current image corresponding to the shooting screen as an end frame, when receiving a preset recognition operation triggered for the shooting screen of the shooting page; and a second determination sub-module for determining, as a target image, an image whose image resolution satisfies a preset resolution condition from the set of images to be processed.

[0102] In an alternative embodiment, the apparatus comprises: The image capturing device further includes a second prompt module for displaying 'recognition failure hint information' on the capturing page when an image satisfying the preset resolution condition is not determined from the set of images to be processed.

[0103] In an image processing device according to an embodiment of the present disclosure, when a preset recognition operation triggered for the captured image of the captured page is received, a target image whose image resolution satisfies a preset resolution condition is determined based on the captured image of the captured page. The target image is then transmitted to the image recognition service for recognition. As described above, before performing image recognition, the image processing device according to an embodiment of the present disclosure first determines a target image whose image resolution satisfies a preset resolution condition, and allows the image recognition service to recognize the target image based on the target image whose resolution satisfies the condition, thereby improving the accuracy of image recognition. Furthermore, because the recognized target image satisfies the preset resolution condition, the occurrence of invalid inputs and invalid recognition is reduced, improving overall image recognition efficiency.

[0104] In addition to the above method and apparatus, an embodiment of the present disclosure further provides a computer-readable storage medium having instructions stored thereon that, when executed by a terminal device, cause the terminal device to implement the image processing method described in the embodiment of the present disclosure.

[0105] An embodiment of the present disclosure further provides a computer program product, which includes computer programs / instructions that, when executed by a processor, implement the image processing method described in the embodiment of the present disclosure.

[0106] Furthermore, an embodiment of the present disclosure further provides an image processing device, which may include a processor 601, a memory 602, an input device 603, and an output device 604, as shown in Fig. 6. The number of processors 601 in the image processing device may be one or more, and Fig. 6 illustrates an example in which one processor is provided. In some embodiments of the present disclosure, the processor 601, the memory 602, the input device 603, and the output device 604 may be connected by a bus or other method, and Fig. 6 illustrates an example in which they are connected via a bus.

[0107] The memory 602 can be used to store software programs and modules, and the processor 601 executes the software programs and modules stored in the memory 602 to perform various functional applications and data processing of the image processing device. The memory 602 may mainly include a program storage area and a data storage area, and the program storage area can store an operating system, an application program required for at least one function, etc. The memory 602 may also include a high-speed random access memory and may further include at least one non-volatile memory such as a magnetic disk storage device, a flash memory device, or other volatile solid-state storage device. The input device 603 can be used to receive input numeric or character information and to generate signal input related to user settings and function control of the image processing device.

[0108] Specifically, in this embodiment, the processor 601 loads executable files corresponding to the processes of one or more application programs into the memory 602 in accordance with the following instructions, and executes the application programs stored in the memory 602 by the processor 601, thereby realizing various functions of the image processing device.

[0109] As should be explained, in this specification, relational terms such as "first," "second," and the like are merely used to distinguish one entity or operation from another and do not necessarily require or imply any actual relationship or ordering between those entities or operations. Furthermore, the terms "comprise," "comprises," or any other variations thereof are intended to cover a non-exclusive inclusion, whereby a process, method, article, or device that includes a set of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent in such process, method, article, or device. Unless otherwise specified, an element qualified by the phrase "comprises one ..." does not exclude the presence of other identical elements in the process, method, article, or device that includes that element.

[0110] The foregoing are merely specific embodiments of the present disclosure, intended to enable those skilled in the art to understand or implement the present disclosure. Various modifications to these examples will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other examples without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the examples described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. 1. An image processing method, comprising: When receiving a preset recognition operation triggered for a photographing screen of a photographing page, determining a target image whose image resolution satisfies a preset resolution condition based on the photographing screen of the photographing page, and determining whether the image resolution satisfies the preset resolution condition using a threshold value of a preset image resolution score; transmitting the target image to an image recognition service for recognizing the target image; Before the step of transmitting the target image to an image recognition service side, extracting image features of the target image at a client; matching the image features with feature parameters of pre-stored preset objects at the client; and when it is determined that the matching is successful, determining in the client a preset object corresponding to the successfully matched feature parameters as an initial recognition result of the target image; In response to this, the step of transmitting the target image to an image recognition service side includes: an image recognition service side including a plurality of image recognition models for recognizing different objects, each of which is configured to recognize a different object; and an image recognition service side including the image recognition service side, which includes the target image and the initial recognition result of the target image at the client, and which calls the corresponding image recognition model based on the initial recognition result transmitted from the client and transmits the image recognition model to the image recognition service side for recognizing the target image.

2. The step of transmitting the target image to an image recognition service side includes: The image processing method according to claim 1, further comprising a step of transmitting the target image to the image recognition service side when it is determined that matching between the image features and the feature parameters of a pre-stored preset object has failed.

3. When receiving a preset recognition operation triggered for a photographing screen of a photographing page, the step of determining a target image whose image resolution satisfies a preset resolution condition based on the photographing screen of the photographing page includes: When receiving a preset recognition operation triggered for a photographing screen of a photographing page, obtaining a current image corresponding to the photographing screen, and determining whether the image resolution of the current image satisfies a preset resolution condition; 2. The image processing method according to claim 1, further comprising a step of determining the current image as a target image if it is determined that the image resolution of the current image satisfies the preset resolution condition, and otherwise acquiring a current image corresponding to the shooting screen and re-executing the step of determining whether the image resolution of the current image satisfies the preset resolution condition until it is determined that the number of current preset image acquisition times has been reached.

4. The image processing method includes:

4. The image processing method according to claim 3, further comprising a step of displaying recognition failure hint information on the photographing page when it is determined that the current preset number of image acquisitions has been reached and an image satisfying the preset resolution condition has not been determined.

5. When receiving a preset recognition operation triggered for a photographing screen of a photographing page, the step of determining a target image whose image resolution satisfies a preset resolution condition based on the photographing screen of the photographing page includes: When receiving a preset recognition operation triggered for a photographing screen of a photographing page, obtaining an image set to be processed including consecutive multi-frame images with a current image corresponding to the photographing screen as an end frame; 2. The image processing method according to claim 1, further comprising the step of determining, as a target image, an image whose image resolution satisfies a preset resolution condition from the set of images to be processed.

6. The image processing method includes: The image processing method according to claim 5 , further comprising a step of displaying recognition failure hint information on the photographing page when an image satisfying the preset resolution condition is not determined from the image set to be processed.

7. The image processing method includes: The image processing method according to claim 1 , further comprising the step of receiving a recognition result transmitted from the image recognition service side and / or a template recommended based on the recognition result.

8. An image processing device, a first determination module for determining a target image whose image resolution satisfies a preset resolution condition based on the captured image of the captured page when receiving a preset recognition operation triggered for the captured image of the captured page, the first determination module using a threshold value of a preset image resolution score to determine whether the image resolution satisfies the preset resolution condition; a transmission module for transmitting the target image to an image recognition service side for recognizing the target image; The image processing device includes: an extraction module at a client for extracting image features of the target image; a matching module for matching the image features with feature parameters of pre-stored preset objects at the client; and a second determination module for determining, when it is determined that the matching is successful, a preset object corresponding to the successfully matched feature parameters in the client as an initial recognition result of the target image; In response, the transmitting module: An image processing device characterized in that the client includes a first transmission submodule for transmitting the target image and the initial recognition result of the target image to an image recognition service side that includes a plurality of image recognition models for recognizing different objects, and for calling the corresponding image recognition model based on the initial recognition result transmitted from the client and transmitting the image recognition service side for recognizing the target image.

9. A computer-readable storage medium having instructions stored therein, the instructions causing a terminal device to realize the image processing method according to any one of claims 1 to 7, when executed by the terminal device.

10. An apparatus comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, the apparatus implementing the image processing method according to any one of claims 1 to 7 when the processor executes the computer program.

11. A computer program comprising computer programs / instructions which, when executed by a processor, implement the image processing method of any one of claims 1 to 7.

Images