Image detection method, device, equipment and medium

Abstract

The embodiment of the invention provides an image detection method and device, equipment and a medium. The method comprises the steps of obtaining a to-be-detected image; intercepting a central area in the to-be-detected image to obtain a target image; performing image denoising on the target image to obtain a denoised image; performing binarization processing on the denoised image to obtain a binarization processing result; and based on the binarization processing result, judging whether the to-be-detected image is an image of blank paper or not. According to the embodiment of the invention, the central area in the to-be-detected image can be intercepted to obtain the target image, the target image is subjected to image denoising, and then whether the to-be-detected image is an image of blank paper or not can be judged based on the denoised image. The interference of the edge area of the to-be-detected image and noise in the to-be-detected image on image detection can be avoided, so that whether the to-be-detected image is an image of blank paper or not can be accurately detected.

Description

Technical Field

[0001] The embodiments of the present specification relate to the field of computer technology, and in particular to an image detection method, device, equipment and medium. Background Art

[0002] With the continuous development of image processing technology, the application of scanners and other equipment is becoming more and more widespread. Scanners can convert paper documents into image formats, making it easier to store and manage documents.

[0003] When using a scanner, a user may input a blank paper into the scanner, such as inputting a stack of documents mixed with blank paper into the scanner. For a blank paper of a blank document, the scanner will scan a blank image. A blank image will affect the user's acquisition efficiency for non-blank images. Therefore, it is necessary to detect whether the image is an image of a blank paper. Summary of the invention

[0004] In view of this, an embodiment of the present specification provides an image detection method. One or more embodiments of the present specification also relate to an image detection device, a computing device, a computer-readable storage medium, and a computer program product to accurately detect whether an image is an image of a blank paper.

[0005] According to a first aspect of an embodiment of this specification, there is provided an image detection method, including: Acquire the image to be detected; Cutting out the central area of ​​the image to be detected to obtain a target image; Performing image denoising on the target image to obtain a denoised image; Performing binarization processing on the denoised image to obtain a binarization processing result; Based on the binarization result, it is determined whether the image to be detected is an image of a blank paper.

[0006] According to a second aspect of the embodiments of this specification, there is provided an image detection device, including: An image acquisition module is configured to acquire an image to be detected; An image capture module is configured to capture a central area of ​​the image to be detected to obtain a target image; An image denoising module is configured to perform image denoising on the target image to obtain a denoised image; A binarization processing module is configured to perform binarization processing on the denoised image to obtain a binarization processing result; The image detection module is configured to determine whether the image to be detected is an image of blank paper based on the binarization processing result.

[0007] According to a third aspect of an embodiment of this specification, a computing device is provided, including: Memory and processor; The memory is used to store computer programs or instructions, and the processor is used to execute the computer programs or instructions. When the computer programs or instructions are executed by the processor, the steps of the above-mentioned image detection method are implemented.

[0008] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores a computer program or instructions, and when the computer program or instructions are executed by a processor, the steps of the above-mentioned image detection method are implemented.

[0009] According to a fifth aspect of the embodiments of this specification, a computer product is provided, including a computer program or instructions, which implement the steps of the above-mentioned image detection method when executed by a processor.

[0010] At least one embodiment of the present specification achieves the following beneficial effects: the embodiments of the present specification can obtain an image to be detected; intercept the central area of ​​the image to be detected to obtain a target image; perform image denoising on the target image to obtain a denoised image; perform binarization processing on the denoised image to obtain a binarization processing result; based on the binarization processing result, determine whether the image to be detected is an image of blank paper. Among them, by using the intercepted central area of ​​the image to be detected as the target image for detecting blank paper, the interference of the edge area of ​​the image to be detected on the detection of blank paper is avoided. In addition, by denoising the target image, the detection accuracy of the blank paper image can be further improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 This is a schematic diagram of an application scenario of an image detection method provided by an embodiment of this specification; Figure 2 is a flow chart of an image detection method provided by an embodiment of this specification; Figure 3 is a schematic diagram of a scanned image with black spots provided by an embodiment of the present specification; Figure 4 It is a front view schematic diagram of a work paper provided by an embodiment of this specification; Figure 5 It is a schematic diagram of a scanned image with words visible on the back provided by an embodiment of this specification; Figure 6 is a schematic diagram of a scanned image of black line noise provided by an embodiment of the present specification; Figure 7is a flowchart of a scanned image detection method in a job submission scenario provided by an embodiment of this specification; Figure 8 is a structural schematic diagram of an image detection device provided by an embodiment of this specification; Fig. 9 It is a structural block diagram of a computing device provided by an embodiment of this specification. DETAILED DESCRIPTION

[0012] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0013] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0014] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0015] In addition, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0016] Figure 1 FIG. 1 is a schematic diagram of an application scenario of an image detection method provided by an embodiment of this specification. Figure 1As shown, at least part of the document 100 to be scanned may be a document containing data content. The user may input the document 100 to be scanned into the scanner 1100. After the scanner 1100 scans the document 100, an image 120 containing data content, a blank image 130 and an image 140 containing data content may be obtained.

[0017] In practical applications, blank images will affect the efficiency of users in obtaining non-blank images. For example, in the scenario of submitting homework or test papers, the machine or human needs to correct the scanned homework images, or grade the scanned test paper images, and the blank image 320 will affect the efficiency of the machine or human in obtaining valid homework images or valid test paper images; in the scenario of book digitization, the blank image 320 will affect the efficiency of readers in reading the scanned book images, etc. Therefore, it is necessary to detect whether the image is an image of blank paper.

[0018] In this specification, an image detection method is provided. This specification also relates to an image detection device, a computing device, a computer-readable storage medium and a computer program product, which are described in detail one by one in the following embodiments.

[0019] Figure 2 This is a flow chart of an image detection method provided by an embodiment of this specification. From a program perspective, the execution subject of the process can be a program installed on a server. From a hardware perspective, the execution subject of the process can be a server capable of performing image detection. The method can specifically include the following steps.

[0020] Step 202: Acquire an image to be detected.

[0021] In the embodiment of the present specification, the image to be detected may be an image used to detect whether it is a blank paper. The image of a blank paper may represent an image that does not contain text content.

[0022] For example, the image to be detected can be a scanned image obtained by a scanner. In scenarios such as homework or test paper submission, book digitization, etc., by detecting whether the scanned image is an image of blank paper, a valid image, i.e., an image of non-blank paper, can be output, thereby improving the efficiency of obtaining valid images. Alternatively, in a scenario where the image of blank paper is used as a document separation mark, by detecting whether the scanned image is an image of blank paper, it can be determined whether the scanned image is a document separation mark, thereby facilitating document processing.

[0023] For another example, the image to be detected may be a sample image in a training sample for training an image processing model. By detecting whether the sample image is an image of blank paper, the image of blank paper may be removed from the training sample, thereby improving the quality of the training sample and improving the performance of the image processing model.

[0024] For another example, the image to be detected may be an image to be identified for a reading target captured by a blind reader. By detecting whether the image to be identified is an image of a blank paper, the user can be prompted whether the reading target contains reading information, thereby improving the user's reading efficiency.

[0025] As an implementation method, the execution subject of the process can obtain the image to be detected from a scanner, a training platform for training an image processing model, or a blind reader.

[0026] Optionally, the image to be detected may be a color image, a grayscale image, or a black-and-white image.

[0027] Optionally, the image to be detected may be in a format such as JPEG, PNG, GIF, TIFF, BMP, SVG, or RAW.

[0028] Step 204: intercepting the central area of ​​the image to be detected to obtain a target image.

[0029] In practical applications, noise may exist in the edge area of ​​the image to be detected. For example, for a scanned image obtained by scanning with a scanner, the scanned image may be torn or folded during the scanning process, resulting in black patches in the edge area of ​​the scanned image.

[0030] Figure 3 FIG. 1 is a schematic diagram of a scanned image with black spots provided by an embodiment of the present specification. Figure 3 As shown, it can be seen that there are black patches 302 in the edge area of ​​the scanned image 300. The presence of the black patches 302 will affect the accuracy of image detection. For example, in an image scene where the number or proportion of black pixels in the image is used to detect whether the image is a blank paper, since the scanned image 300 contains the black patches 302, the black patches 302 will be identified as black pixels, and the scanned image 300 may be detected as an image of a non-blank paper.

[0031] The embodiments of the present specification can capture the central area of ​​the image to be detected to obtain a target image, and determine whether the image to be detected is an image of blank paper by detecting the target image, thereby avoiding interference of the edge area of ​​the image to be detected on the detection result and improving the accuracy of image detection.

[0032] In practical applications, the target image can be obtained by intercepting the central area of ​​the image to be detected based on the central point of the image to be detected. The target image can include the central point.

[0033] As a specific implementation, the center point of the image to be detected can be used as the center point of the target image, and the center area of ​​the image to be detected can be intercepted according to a preset ratio to obtain the target image. The preset ratio can represent the size ratio of the target image in the image to be detected.

[0034] Optionally, the preset ratio may be 80%, 85%, 90%, 95%, 99%, etc., which is not limited here.

[0035] In the embodiment of this specification, the target image can be used to determine whether the image to be detected is an image of a blank paper. Since the effective information in the image is generally located in the central area of ​​the image, by detecting the target image, it is possible to accurately determine whether the image to be detected is an image of a blank paper.

[0036] In the embodiment of the present specification, the target image may be a partial image in the image to be detected. By detecting at least a portion of the image to be detected to determine whether the image to be detected is an image of a blank paper, the image detection efficiency can be improved compared to detecting all portions of the image to be detected.

[0037] Step 206: performing image denoising on the target image to obtain a denoised image.

[0038] In practical applications, the target image may have noise. For example, for a scanned image, since the pixels in the scanned image are obtained by the scanner through electronic devices to perform information quantization sampling, the pixels in the scanned image may be superimposed with electronic noise and quantization noise. For example, randomly distributed bright or dark spots may appear in the scanned image, vertical or horizontal bright or dark lines may appear, and details may be lost such as local blur or unclear edges.

[0039] In addition, there may be other noises in the scanned image.

[0040] Figure 4 1 is a front view of a work paper provided by an embodiment of this specification. Figure 4 As shown, the text 402 written by the student on the front side 400 of the homework paper can be seen. In actual applications, when students write homework, if they write too hard or the homework paper is thin, the written text may be visible from the back side of the homework paper. When scanning the homework paper with a scanner, if the back side of the homework paper is mistakenly input into the scanner, or in the double-sided scanning mode, a scanned image with visible text on the back side may be obtained. For this type of scanned image, the visible text is noise.

[0041] Figure 5 FIG. 1 is a schematic diagram of a scanned image with transparent characters on the back provided by an embodiment of this specification. Figure 5As shown, the scanned image 500 on the back side of the work paper has a noise of transparent characters 502 . Figure 5 The font of the transparent word 502 in FIG. 5 is expressed in black, just to better show the noise of the transparent word 502. In actual applications, the transparent word 502 may be gray or other colors.

[0042] Black line noise may appear due to scanning errors and other reasons during the scanning process of the paper to be scanned. Figure 6 FIG. 1 is a schematic diagram of a scanned image of black line noise provided by an embodiment of the present specification. Figure 6 As shown, black line noise 602 can be seen in the scanned image 600 .

[0043] In practical applications, various denoising methods can be used for image denoising. For example, at least one of the image morphological processing methods such as erosion, dilation, opening, closing, morphological gradient, top-hat transform and bottom-hat transform can be used for image denoising.

[0044] Optionally, other image denoising methods may be used, such as at least one of mean filtering, median filtering, Gaussian filtering, bilateral filtering, wavelet denoising, Fourier transform denoising, and deep learning models, etc. to perform image denoising.

[0045] In the embodiments of this specification, the target image can be subjected to image denoising to obtain a denoised image, and then image detection can be performed based on the denoised image, thereby improving the accuracy of image detection.

[0046] Step 208: performing binarization processing on the denoised image to obtain a binarization processing result.

[0047] The binarization process can convert the pixels in the image into black or white, thereby obtaining a binarized image containing only black pixels and white pixels. The binarization process result of the embodiment of the present specification may include a binarized image.

[0048] In practical applications, if the image to be detected is a color image, the denoised image can be first converted into a grayscale image, and then the grayscale image can be binarized.

[0049] Specifically, in the process of binarizing the grayscale image, the grayscale value of each pixel in the grayscale image can be compared with a preset threshold. If the comparison result is that the grayscale value is greater than or equal to the preset threshold, the pixel corresponding to the grayscale value can be set to a white pixel. If the comparison result is that the grayscale value is less than the preset threshold, the pixel corresponding to the grayscale value can be set to a black pixel.

[0050] In the embodiments of the present specification, at least one of a global thresholding method, an adaptive thresholding method, an Otsu method, and a multi-thresholding method may be used to determine the preset threshold.

[0051] Optionally, if the image to be detected is a grayscale image, binarization processing can be performed directly.

[0052] Optionally, if the image to be detected is a black and white image, step 208 may be omitted.

[0053] Step 210: Based on the binarization result, determine whether the image to be detected is an image of blank paper.

[0054] In the embodiment of the present specification, the binarization result may be a binarized image. Based on the binarized image, it may be determined whether the image to be detected is an image of a blank paper.

[0055] Specifically, whether the image to be detected is an image of blank paper can be determined based on the number of white pixels and the number of black pixels in the binary image. For example, the number of white pixels can be counted, and if the proportion of the number of white pixels in the binary image is greater than or equal to a preset proportion threshold, it is determined that the image to be detected is an image of blank paper.

[0056] The embodiment of the specification can capture the central area of ​​the image to be detected to obtain the target image, perform image denoising on the target image, and then determine whether the image to be detected is an image of blank paper based on the denoised image. Since the interference of the edge area of ​​the image to be detected and the interference of noise in the image to be detected are avoided, image detection can be performed accurately.

[0057] It should be understood that the order of some steps in the methods described in one or more embodiments of this specification can be interchanged according to actual needs, such as the execution order of step 202, step 204 and step 206 can be interchanged, or some of the steps can be omitted or deleted.

[0058] based on Figure 2The method, the examples of this specification also provide some specific implementation plans of the method, which are described below.

[0059] In practical applications, the image to be detected can be acquired by a scanner, a high-speed scanner or a camera. The camera can be a camera installed on a terminal. The terminal can be a mobile phone, a reader for the blind, a desktop computer, a laptop computer, a smart phone, or one or more of a desktop computer, a laptop computer, a tablet computer, an Internet of Things device, a portable wearable device or an immersive image display device.

[0060] In the embodiments of this specification, the central area of ​​the image to be detected can be intercepted based on the central point of the image to be detected to obtain the target image. Optionally, intercepting the central area of ​​the image to be detected to obtain the target image may specifically include: Determine the center point of the image to be detected.

[0061] Based on the center point, a center area in the image to be detected is intercepted according to a preset interception ratio to obtain a target image; the target image includes the center point.

[0062] In the embodiment of this specification, the center point of the image to be detected can be the geometric center of the image to be detected. Specifically, for example, if the image to be detected is a rectangle, the intersection of the diagonals of the rectangle can be used as the center point of the image to be detected. Alternatively, the center point can be determined based on the coordinates of the vertices of the rectangle. For example, the coordinates of the vertices of the rectangle are: , , , , then the coordinates can be The corresponding point is taken as the center point of the image to be detected.

[0063] Optionally, the center point of the image to be detected can be used as the center point of the target image, and the center area of ​​the image to be detected can be intercepted according to a preset interception ratio to obtain the target image. The preset interception ratio is the size ratio of the target image in the image to be detected.

[0064] Specifically, the size of the image to be detected can be calculated first, and then the size of the target image can be calculated based on a preset interception ratio, and then the target image can be intercepted using the center point of the image to be detected as the center point of the target image.

[0065] As a specific implementation, the center point of the image to be detected may not be used as the center point of the target image. Instead, the center point of the image to be detected is used as a reference, and the center point of the image to be detected is separated from the center point of the target image by a preset distance, and the area including the center point of the image to be detected is intercepted to obtain the target image.

[0066] In practical applications, the center point of the image to be detected may also be determined. For example, the target image may be obtained based on the coordinates of each vertex of the image to be detected. Optionally, the central area of ​​the image to be detected may be intercepted to obtain the target image, which may specifically include: Based on the coordinates of each vertex of the image to be detected and a preset interception ratio, the coordinates of each vertex of the target image are determined.

[0067] Based on the coordinates of each vertex of the target image, a central area in the image to be detected is intercepted to obtain the target image.

[0068] In the embodiment of the present specification, the product of the vertex coordinates of the image to be detected and the preset interception ratio can be used as the vertex coordinates of the target image.

[0069] Based on the coordinates of each vertex of the target image, in the process of intercepting the central area of ​​the image to be detected, adjacent vertices in the target image can be connected to obtain each boundary of the target image. Then, the target image is intercepted from the image to be detected with reference to each boundary of the target image.

[0070] In practical applications, the target image can also be obtained based on the boundary of the image to be detected. Optionally, the method of intercepting the central area of ​​the image to be detected to obtain the target image may specifically include: Any boundary of the image to be detected is moved to an opposite boundary of any boundary by a preset distance to obtain a target image with a preset interception ratio.

[0071] Specifically, the preset distance may be less than or equal to half of the size of the adjacent boundary of any boundary in the image to be detected. For example, if the image to be detected is a rectangular image with a length of 10 cm and a width of 8 cm, the long side of the rectangular image may be moved 2 cm toward the other long side of the rectangular image; or the wide side of the rectangular image may be moved 3 cm toward the other wide side of the rectangular image, etc.

[0072] In practical applications, one boundary of the image to be detected may be moved, or multiple boundaries of the image to be detected may be moved, etc., which is not limited here.

[0073] For example, for a boundary of a moving image to be detected, a closed rectangle formed by the moved boundary and the unmoved boundary can be used as the target image.

[0074] For another example, for all the boundaries of the image to be detected that are moved, a closed rectangle formed by all the moved boundaries can be used as the target image.

[0075] In the embodiments of this specification, the shape of the target image can be the same as the shape of the image to be detected. For example, if the image to be detected is a rectangle, the target image can be a rectangle. For example, if the image to be detected is a circle, the target image can be a circle, etc.

[0076] Optionally, the shape of the target image may be different from the shape of the image to be detected. For example, if the image to be detected is rectangular, the target image may be circular or rhombus-shaped. For example, if the image to be detected is circular, the target image may be rectangular or rhombus-shaped.

[0077] In the present embodiment, the preset interception ratio may be 80%, 85%, 90%, 95%, 99%, etc. In practical applications, it may be determined according to actual needs and is not limited here.

[0078] In the embodiment of this specification, the dilation operation in the image morphological processing method can be used to denoise the image. Optionally, the image denoising of the target image to obtain the denoised image can specifically include: Perform image expansion on the target image to obtain an expanded image.

[0079] In practical applications, the specific process of performing image expansion on a target image may include the following steps.

[0080] Initialize the expansion kernel. The expansion kernel is also called a structural element, which is used to represent the shape and size of the expansion operation. The expansion kernel is usually represented by a matrix, and the element value of the matrix can be 0 or 1. Specifically, the shape of the expansion kernel can be circular, elliptical, rectangular, etc. The size of the expansion kernel can be 1×1 pixels, 3×3 pixels, 5×5 pixels, 7×7 pixels, etc. Optionally, in the embodiments of this specification, the shape of the expansion kernel can be set to a circle and the size can be set to 3×3 pixels.

[0081] In practical applications, the anchor point of the expansion kernel can also be determined. The anchor point is a reference point in the expansion kernel, which is used to indicate the directionality of the expansion operation. In the embodiments of this specification, the center point of the expansion kernel can be used as the anchor point.

[0082] Optionally, each pixel point in the target image may be traversed based on the dilation kernel. Specifically, the anchor point of the dilation kernel may be placed on each pixel point in the target image in sequence to obtain the overlapping area of ​​the target image with the dilation kernel.

[0083] If at least one pixel in the overlapping area is a foreground pixel, the pixel at the position where the anchor point is placed is set as a foreground pixel. The foreground pixel can be understood as a white pixel. Setting the pixel at the position where the anchor point is placed as a foreground pixel can be setting the pixel value of the pixel at the position where the anchor point is placed to a pixel value representing a white pixel, such as 1 or 255.

[0084] If there are no foreground pixels in the overlapping area, that is, the overlapping area is all background pixels, the pixels at the position where the anchor point is placed will not be changed. Specifically, the pixel value of the pixel at the position where the anchor point is placed will not be changed. The background pixel can be understood as a black pixel, and the pixel value of the background pixel can be 0.

[0085] In practical applications, the pixels in the target image can be traversed multiple times based on the dilation kernel, such as 2 times, 3 times, 5 times, etc. After the dilation is completed, the image denoising is completed, and the obtained dilated image can be the denoised image.

[0086] In practical applications, the object in the target image may have holes or gaps. For example, the target image may be an image of text, and the text in the target image may have small holes, or there may be gaps between the texts. The existence of these holes and gaps may cause the text in the subsequent binarization processing results to be incomplete, thereby affecting the image detection results. The embodiments of this specification can fill holes and gaps based on the dilation operation to ensure the accuracy of image detection.

[0087] Furthermore, the boundaries of the text in the target image may be broken or discontinuous due to noise, etc. The dilation operation can be used to expand the boundaries of the text, thereby smoothing the boundaries of the text and making the boundaries of the text more continuous and complete, thereby reducing the impact of noise on the boundaries of the text and improving the accuracy of image detection.

[0088] Furthermore, by dilating the target image, the number of white pixels can be increased, thereby enhancing the foreground area in the target image (the foreground area is the area containing white pixels). In the subsequent binarization process, it is easier to distinguish the foreground area and the background area of ​​the target image (the background area is the area containing black pixels), thereby improving the accuracy of the binarization process and further improving the accuracy of image detection.

[0089] In the embodiment of the specification, the denoised image may be binarized. The binarization of the denoised image may include: Obtaining a global threshold or an adaptive threshold corresponding to the denoised image; Based on the global threshold or the adaptive threshold, the denoised image is binarized.

[0090] In the embodiments of this specification, the threshold corresponding to the denoised image may be a global threshold. Binarization of the denoised image based on the global threshold does not require complex threshold calculation or region segmentation, which is simple and efficient.

[0091] Optionally, the global threshold value range can be .

[0092] As a specific implementation, the threshold corresponding to the denoised image may also be an adaptive threshold. The adaptive threshold may indicate that different regions in the denoised image may correspond to different thresholds. This can adapt to local changes in the denoised image. For example, for regions with different brightness or contrast, different thresholds may be used for regional compensation to ensure the accuracy of the binarization processing result, thereby ensuring the accuracy of image detection.

[0093] Optionally, the denoised image may be divided into different regions, and for any region in the denoised image, the adaptive threshold of any region may be obtained by the following calculation formula:

[0094] in, represents the adaptive threshold of any area in the denoised image, represents the average gray value of any region, Indicates the preset scale factor, , Represents the standard deviation of the grayscale values ​​of each pixel in any area.

[0095] In the embodiment of the present specification, whether the image to be detected is an image of a blank paper can be judged based on the binarization processing result. Optionally, the binarization processing result is a binarized image; judging whether the image to be detected is an image of a blank paper based on the binarization processing result may specifically include: A first number of white pixels and a second number of black pixels in the binary image are counted.

[0096] Calculate the proportion of the first quantity or the second quantity in the binary image.

[0097] Determine whether the proportion is greater than or equal to a preset proportion threshold.

[0098] If the proportion is greater than or equal to the preset proportion threshold, it is determined that the image to be detected is an image of blank paper.

[0099] In practical applications, the image of a blank paper may be an image of a white paper. Or the image of a blank paper may be an image of a paper that is more white than black. Compared to black, a color that is more white may refer to a color whose value of at least one of the RGB channels is close to 255. Close to 255 can be understood as a difference from 255 that is within a preset difference range. For example, these colors may be beige, silver, and the like. For the paper of the above colors, the user may use a black pen or a pen that is more black to write text. If the number of white pixels in the binarized image of the image of the above color paper is greater than or equal to a preset percentage threshold in the binarized image, the image of the above color paper may be considered to be an image of a blank paper.

[0100] As a specific implementation, the image of a blank paper may also be an image of a black paper. For example, an image of a black A4 paper. Or the image of a blank paper may also be an image of a paper that is more black than white. Compared to white, a color that is more black may refer to a color whose value of at least one channel in the RGB channels is close to 0. Close to 0 can be understood as a difference from 0 within a preset difference range. For example, these colors may be dark gray, dark brown, and the like. For the above-mentioned color papers, the user may use a white pen or a whitish pen to write text. If the number of black pixels in the binarized image of the image of the above-mentioned color paper is greater than or equal to a preset proportion threshold in the binarized image, the image of the above-mentioned color paper may be considered to be an image of a blank paper.

[0101] In the embodiment of this specification, the value range of the preset proportion threshold is Specifically, the preset percentage threshold may be 99%, 99.5%, or 99.91%, etc., which is not limited here.

[0102] The preset percentage threshold value in the embodiments of this specification is relatively large. For example, for a white paper image, even if the white paper image contains a lot of black text content, due to the gaps between the text and the line spacing, the percentage of black pixels in the white paper image does not reach the preset percentage threshold value set in the embodiments of this specification. Therefore, no matter what color the image of the blank paper is, by calculating the percentage of white pixels or the percentage of black pixels, it can be determined whether the image to be detected is an image of a blank paper.

[0103] Optionally, if the proportion is less than the preset proportion threshold, the image to be detected is determined to be an image of a non-blank paper. For example, for a white paper image, or a paper image that is more white than black, if the first number of white pixels is less than the preset proportion threshold, it can be indicated that there is text content in the image, and the image is determined to be an image of a non-blank paper. The same is true for a black paper image, or a paper image that is more black than white, and will not be elaborated here.

[0104] Specifically, when the proportion is less than the preset proportion threshold, if the proportion is greater than 1 minus the preset threshold, it is determined that the image to be detected is an image of a non-blank paper. Optionally, if 1-preset proportion threshold < proportion < preset proportion threshold, it can be determined that the image to be detected is an image of a non-blank paper.

[0105] In practical applications, the binary image corresponding to the image of a blank paper may generally contain only white pixels or only black pixels. For example, the binary image corresponding to a white A4 paper may contain only white pixels, and the binary image corresponding to a black A4 paper may contain only black pixels, etc. The binary image corresponding to an image of a non-blank paper may contain both white pixels and black pixels. Therefore, it is possible to detect whether the image to be detected is an image of a blank paper by determining whether the number of white pixels or black pixels in the binary image is zero. Optionally, determining whether the image to be detected is an image of a blank paper based on the binarization processing result may specifically include: Counting a first number of white pixels and a second number of black pixels in the binary image; If the first number is zero or the second number is zero, it is determined that the image to be detected is an image of a blank paper.

[0106] In practical applications, the image to be detected may be an image with transparent text on the back, such as Figure 5 For an image with transparent characters on the back, based on the binarization result, it may be detected that the image to be detected is an image of a non-blank paper, resulting in a detection error. In order to avoid detection errors, the method provided in the embodiment of this specification may optionally further include: The image to be detected is mirror-flipped to obtain a mirror-flipped image.

[0107] The optical character recognition technology is used to perform text recognition on the mirror-flipped image to obtain a recognition result.

[0108] If the recognition result indicates that text exists in the image after the mirror flipping, it is determined that the image to be detected is an image of a blank paper.

[0109] Determining that the image to be detected is an image of a non-blank paper may specifically include: If the recognition result indicates that there is no text in the image after the mirror flipping, it is determined that the image to be detected is an image of a non-blank paper.

[0110] Optical Character Recognition (OCR) is a technology that converts text in an image into text that can be processed by a computer. Optical Character Recognition is often used for text recognition.

[0111] The optical character recognition technology in the embodiments of this specification can recognize text in standard fonts in an image. Text in standard fonts can refer to fonts written in a standard writing method. The optical character recognition technology in the embodiments of this specification cannot recognize inverted fonts, mirrored fonts, etc.

[0112] The embodiment of this specification can mirror-flip the image to be detected to obtain a mirror-flipped image. Then, the optical character recognition technology can be used to perform text recognition on the mirror-flipped image. If the recognition result indicates that there is text in the mirror-flipped image, it can be considered that the image to be detected is an image of a paper with words on the back, and the image to be detected is an image of a non-blank paper. If the recognition result indicates that there is no text in the mirror-flipped image, it can be considered that the image to be detected is an image of a blank paper.

[0113] The embodiment of the present specification can improve the detection accuracy of the image of paper with transparent words on the back by mirror-flipping the image to be detected and further identifying the image after mirror-flipping.

[0114] In practical applications, the noise in the image to be detected may be randomly distributed in the image to be detected. For example, electronic noise and quantization noise. However, the text may not be randomly distributed in the image to be detected. Therefore, based on the grayscale distribution of the grayscale image of the image to be detected, it can be detected whether the image to be detected is an image of blank paper. Optionally, the method of the embodiment of this specification may also include: Grayscale processing is performed on the image to be detected to obtain a grayscale image.

[0115] A grayscale histogram of the grayscale image is determined based on the grayscale value of each pixel in the grayscale image.

[0116] Through the grayscale histogram, a normality test is used to determine whether the grayscale value of each pixel in the grayscale image belongs to a normal distribution.

[0117] If the grayscale value of each pixel in the grayscale image belongs to a normal distribution, it is determined that the image to be detected is an image of a blank paper.

[0118] In the embodiment of this specification, if the image to be detected is a color image, the image to be detected can be grayed. Specifically, the values ​​of the three channels of RGB of each pixel in the image to be detected can be converted into a specific value representing brightness. The specific value can represent the gray level of the pixel from black to white.

[0119] Grayscale histogram is a tool that counts the number of occurrences or frequencies of each grayscale value in a grayscale image and expresses it in a histogram. Grayscale histogram can represent the distribution of different grayscale values ​​in a grayscale image.

[0120] In the embodiment of the present specification, the horizontal axis of the grayscale histogram may represent the grayscale value, and the vertical axis of the grayscale histogram may represent the number of occurrences of the grayscale value.

[0121] In the embodiments of this specification, the normality test method used may be at least one of the following test methods: KS test (Kolmogorov-Smirnov test), Shapiro-Wilk test, Anderson-Darling test and Lilliefors test.

[0122] Optionally, if the grayscale values ​​of each pixel in the grayscale image belong to a normal distribution, it can be said that the image to be detected contains only noise. If the grayscale values ​​of each pixel in the grayscale image do not belong to a normal distribution, it can be said that the image to be detected contains not only noise but also text. Therefore, if the grayscale values ​​of each pixel in the grayscale image belong to a normal distribution, it is determined that the image to be detected is an image of a blank paper; if the grayscale values ​​of each pixel in the grayscale image do not belong to a normal distribution, it is determined that the image to be detected is an image of a non-blank paper.

[0123] In practical applications, if the number of pixels in the image to be detected is too large, such as greater than a preset number threshold, the detection accuracy may be poor by counting the number of black and white pixels in the binary image of the image to be detected to determine whether the image to be detected is an image of blank paper. Therefore, in order to improve the accuracy of image detection, when the number of pixels in the image to be detected is greater than the preset number threshold, the grayscale histogram can be used to determine whether the grayscale value of each pixel in the grayscale image of the image to be detected belongs to a normal distribution to determine whether the image to be detected is an image of blank paper. When the number of pixels in the image to be detected is less than or equal to the preset number threshold, the number of black and white pixels in the binary image of the image to be detected is counted to determine whether the image to be detected is an image of blank paper.

[0124] In practical applications, the binarization result and the distribution of the grayscale values ​​of each pixel in the grayscale image can also be combined to determine whether the image to be detected is an image of blank paper. The method can also include: If the binarization result indicates that the image to be detected is an image of blank paper, and the grayscale value of each pixel in the grayscale image belongs to a normal distribution, it is determined that the image to be detected is an image of blank paper.

[0125] By combining the binarization result and the distribution of the grayscale value of each pixel in the grayscale image to determine whether the image to be detected is an image of blank paper, the accuracy of image detection can be improved.

[0126] Figure 7 FIG. 1 is a flowchart of a scanned image detection method in a job submission scenario provided by an embodiment of the present specification. Figure 7 As shown, the method may further include the following steps.

[0127] Step 702: Acquire an image to be detected.

[0128] In the embodiment of this specification, the image to be detected may be an image used to detect whether it is a blank paper. For example, the image to be detected may be a scanned image obtained by a scanner, such as a job scanned image.

[0129] Specifically, students can scan the completed homework papers, and then send the homework scanned images to teachers for review or homework correction, etc. Alternatively, teachers can collect the homework papers completed by students, and scan the homework papers to obtain the homework scanned images, etc.

[0130] The homework scanned images may be multiple homework scanned images corresponding to multiple students. One student may correspond to one or more homework scanned images, etc.

[0131] Step 704: Grayscale processing.

[0132] In the embodiment of this specification, the server can perform grayscale processing on the job scan image.

[0133] As a specific implementation, if the image to be detected is a grayscale image, the image to be detected may not be grayscaled. Step 704 may be omitted.

[0134] Step 706: intercept the central area.

[0135] In actual applications, the server can intercept the central area of ​​the job scan image based on the central point of the job scan image according to a preset interception ratio to obtain the target image. The preset interception ratio is the size ratio of the target image in the image to be detected.

[0136] In the embodiment of this specification, the execution order of step 706 is not limited, as long as it is executed before step 712. For example, it can be executed after step 702 and before step 704, or after step 710 and before step 712.

[0137] Step 708: Morphological dilation of the image.

[0138] In the embodiment of this specification, the server may perform image morphological dilation on the target image to remove noise in the target image.

[0139] In practical applications, other denoising methods can also be used, such as corrosion, top-hat transform, mean filtering, Gaussian filtering, wavelet transform denoising, Fourier transform denoising, and image denoising based on deep learning models, etc.

[0140] In the embodiment of this specification, the execution order of step 608 is not limited, as long as it is executed before step 612. For example, it can be executed after step 604 and before step 606, or after step 610 and before step 612.

[0141] Step 710: Binarization processing.

[0142] In the embodiments of the present specification, the server may perform binarization processing on the target image to obtain a binary image containing only black pixels and white pixels.

[0143] Specifically, at least one of a fixed threshold method (Global Thresholding), an adaptive threshold method (AdaptiveThresholding), an Otsu method, and a multi-thresholding method (Multi-thresholding) can be used to determine a corresponding preset threshold in the target image binarization process.

[0144] Step 712: Count the number of black and white pixels.

[0145] In the embodiment of the present specification, the server may count the number of black pixels, the number of white pixels, etc. in the binary image.

[0146] Step 714: Determine whether it is an image of a blank paper.

[0147] Optionally, the server may determine whether the scanned image to be scanned is an image of blank paper based on the number of black pixels and the number of white pixels counted in step 712 .

[0148] For example, the proportion of white pixels in the binary image is calculated. If the proportion is greater than or equal to a preset proportion threshold, the job scan image is determined to be an image of a blank paper. If the proportion is less than the preset proportion threshold, it can be determined that the job scan image is not an image of a blank paper.

[0149] Optionally, the preset percentage threshold value range is . Specifically, it can be 99.5% or 99.92%, etc.

[0150] Step 716: Filter and delete.

[0151] In the embodiment of the present specification, if the job scan image is an image of a blank paper, the job scan image may be filtered and deleted, for example, the image of the blank paper may be deleted from the plurality of job scan images.

[0152] Step 718: Subsequent processing.

[0153] In practical applications, if the scanned image of the homework is not an image of a blank paper, subsequent processing may be performed, such as a teacher marking the homework. In the scenario of machine marking homework, the scanned image of the homework may also be marked by the machine.

[0154] The embodiment of the specification uses the central area of ​​the intercepted homework scan image as the target image for detecting blank paper, thereby avoiding the interference of the edge area of ​​the homework scan image on the blank paper detection. In addition, by performing image morphological expansion on the homework correction image, the noise in the homework correction image can be removed, thereby further improving the detection accuracy of the blank paper image, which is convenient for teachers or machines to correct homework.

[0155] Corresponding to the above-mentioned image detection method embodiment, this specification also provides an image detection device embodiment, Figure 8 FIG. 1 is a schematic diagram of the structure of an image detection device provided by an embodiment of this specification. Figure 8 As shown, the device comprises: The image acquisition module 802 is configured to acquire an image to be detected.

[0156] The image capture module 804 is configured to capture the central area of ​​the image to be detected to obtain a target image.

[0157] The image denoising module 806 is configured to perform image denoising on the target image to obtain a denoised image.

[0158] The binarization processing module 808 is configured to perform binarization processing on the denoised image to obtain a binarization processing result.

[0159] The image detection module 810 is configured to determine whether the image to be detected is an image of blank paper based on the binarization result.

[0160] Optionally, the image capture module 804 may be specifically configured as follows: Determine the center point of the image to be detected.

[0161] Based on the center point, a center area in the image to be detected is intercepted according to a preset interception ratio to obtain a target image; the target image includes the center point.

[0162] Optionally, the image denoising module 806 may be specifically configured as follows: Perform image expansion on the target image to obtain an expanded image.

[0163] Optionally, the binarization processing module 808 may be specifically configured as follows: Obtain a global threshold or an adaptive threshold corresponding to the denoised image.

[0164] Based on the global threshold or the adaptive threshold, the denoised image is binarized.

[0165] Optionally, the value range of the global threshold is ; The adaptive threshold is calculated by the following calculation formula:

[0166] in, represents the adaptive threshold of any area in the denoised image, represents the average gray value of any region, Indicates the preset scale factor, , Represents the standard deviation of the grayscale values ​​of each pixel in any area.

[0167] Optionally, the binarization processing result is a binarized image; the image detection module 810 may be specifically configured as follows: A first number of white pixels and a second number of black pixels in the binary image are counted.

[0168] Calculate the proportion of the first number or the second number in the binary image.

[0169] Determine whether the proportion is greater than or equal to a preset proportion threshold.

[0170] If the proportion is greater than or equal to the preset proportion threshold, it is determined that the image to be detected is an image of blank paper.

[0171] Optionally, the device may further include: If the proportion is less than the preset proportion threshold, it is determined that the image to be detected is an image of a non-blank paper.

[0172] Optionally, the device may further include: Performing mirror flipping on the image to be detected to obtain a mirror flipped image; Using optical character recognition technology to perform text recognition on the mirror-flipped image to obtain a recognition result; If the recognition result indicates that text exists in the image after the mirror flipping, determining that the image to be detected is an image of a blank paper; Determining that the image to be detected is an image of a non-blank paper may specifically include: If the recognition result indicates that there is no text in the image after the mirror flipping, it is determined that the image to be detected is an image of a non-blank paper.

[0173] Optionally, the device may further include: The grayscale module is configured to perform grayscale processing on the image to be detected to obtain a grayscale image.

[0174] The grayscale histogram determination module is configured to determine the grayscale histogram of the grayscale image based on the grayscale value of each pixel in the grayscale image.

[0175] The judgment module is configured to judge whether the grayscale value of each pixel in the grayscale image belongs to a normal distribution through the grayscale histogram and a normality test.

[0176] The image determination module is configured to determine that the image to be detected is an image of blank paper if the grayscale value of each pixel in the grayscale image belongs to a normal distribution.

[0177] Optionally, the image to be detected is acquired by a scanner, a high-speed scanner or a camera.

[0178] The above is a schematic scheme of an image detection device of this embodiment. It should be noted that the technical scheme of the image detection device and the technical scheme of the above-mentioned image detection method belong to the same concept, and if the details of the technical scheme of the image detection device are not described in detail, please refer to the description of the technical scheme of the above-mentioned image detection method.

[0179] Fig. 9 The block diagram of a computing device 900 provided according to an embodiment of the present specification is shown. The components of the computing device 800 include but are not limited to a memory 910 and a processor 920. The processor 920 is connected to the memory 910 via a bus 930, and the database 950 is used to store data.

[0180] The computing device 900 also includes an access device 940 that enables the computing device 900 to communicate via one or more networks 960. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 540 may include one or more of any type of network interface (e.g., a network interface card (NIC)) of wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a world-wide interoperability for microwave access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, and a near field communication (NFC).

[0181] In one embodiment of the present specification, the above components of the computing device 900 and Fig. 9 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Fig. 9 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0182] The computing device 900 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 900 may also be a mobile or stationary server.

[0183] The processor 920 is used to execute the following computer executable instructions, which implement the steps of the above-mentioned image detection method when executed by the processor.

[0184] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the above-mentioned image detection method belong to the same concept, and the details not described in detail in the technical scheme of the computing device can be referred to the description of the technical scheme of the above-mentioned image detection method.

[0185] An embodiment of the present specification further provides a computer-readable storage medium storing computer-executable instructions, which can implement the steps of the above-mentioned image detection method when executed by a processor.

[0186] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the above-mentioned image detection method belong to the same concept, and the details not described in detail in the technical scheme of the storage medium can be referred to the description of the technical scheme of the above-mentioned image detection method.

[0187] An embodiment of the present specification further provides a computer program product, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned image detection method.

[0188] The above is an illustrative solution of a computer program of this embodiment. It should be noted that the technical solution of the computer program and the technical solution of the above-mentioned image detection method belong to the same concept, and the details not described in detail in the technical solution of the computer program can be referred to the description of the technical solution of the above-mentioned image detection method.

[0189] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0190] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0191] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0192] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0193] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can well understand and use this specification. This specification is only limited by the claims and their full scope and equivalents.

Claims

1. An image detection method, characterized in that: include: Acquire the image to be detected; Cutting out the central area of ​​the image to be detected to obtain a target image; Performing image denoising on the target image to obtain a denoised image; Performing binarization processing on the denoised image to obtain a binarization processing result; Based on the binarization result, it is determined whether the image to be detected is an image of a blank paper.

2. The method according to claim 1, characterized in that: The method of intercepting the central area of ​​the image to be detected to obtain the target image specifically includes: Determine the center point of the image to be detected; Based on the center point, a center area in the image to be detected is intercepted according to a preset interception ratio to obtain a target image; the target image includes the center point.

3. The method according to claim 1, characterized in that The performing image denoising on the target image to obtain a denoised image specifically includes: Perform image expansion on the target image to obtain an expanded image.

4. The method according to claim 1, characterized in that: The binarization processing of the denoised image specifically includes: Obtaining a global threshold or an adaptive threshold corresponding to the denoised image; Based on the global threshold or the adaptive threshold, the denoised image is binarized.

5. The method according to claim 4, characterized in that The value range of the global threshold is ; The adaptive threshold is calculated by the following calculation formula: in, represents the adaptive threshold of any area in the denoised image, represents the average gray value of any region, Indicates the preset scale factor, , Represents the standard deviation of the grayscale values ​​of each pixel in any area.

6. The method according to claim 1, characterized in that The binarization result is a binarized image; and judging whether the image to be detected is an image of a blank paper based on the binarization result specifically includes: Counting a first number of white pixels and a second number of black pixels in the binary image; Calculating a proportion of the first quantity or the second quantity in the binary image; Determine whether the proportion is greater than or equal to a preset proportion threshold; If the proportion is greater than or equal to the preset proportion threshold, it is determined that the image to be detected is an image of blank paper.

7. The method according to claim 6, characterized in that Also includes: If the proportion is less than the preset proportion threshold, it is determined that the image to be detected is an image of a non-blank paper.

8. The method according to claim 7, characterized in that Also includes: Performing mirror flipping on the image to be detected to obtain a mirror flipped image; Using optical character recognition technology to perform text recognition on the mirror-flipped image to obtain a recognition result; If the recognition result indicates that text exists in the image after the mirror flipping, determining that the image to be detected is an image of a blank paper; The step of determining that the image to be detected is an image of a non-blank paper specifically includes: If the recognition result indicates that there is no text in the image after the mirror flipping, it is determined that the image to be detected is an image of a non-blank paper.

9. The method according to claim 1, characterized in that: Also includes: Performing grayscale processing on the image to be detected to obtain a grayscale image; Determining a grayscale histogram of the grayscale image based on the grayscale value of each pixel in the grayscale image; By using the grayscale histogram, a normality test is used to determine whether the grayscale value of each pixel in the grayscale image belongs to a normal distribution; If the grayscale value of each pixel in the grayscale image belongs to a normal distribution, it is determined that the image to be detected is an image of a blank paper.

10. The method according to claim 1, characterized in that The image to be detected is acquired by a scanner, a high-speed scanner or a camera.

11. An image detection device, characterized in that: include: An image acquisition module is configured to acquire an image to be detected; An image capture module is configured to capture a central area of ​​the image to be detected to obtain a target image; An image denoising module is configured to perform image denoising on the target image to obtain a denoised image; A binarization processing module is configured to perform binarization processing on the denoised image to obtain a binarization processing result; The image detection module is configured to determine whether the image to be detected is an image of blank paper based on the binarization processing result.

12. A computing device, characterized in that: include: Memory and processor; The memory is used to store computer programs or instructions, and the processor is used to execute the computer programs or instructions. When the computer programs or instructions are executed by the processor, the steps of the image detection method according to any one of claims 1 to 10 are implemented.

13. A computer-readable storage medium, characterized in that: It stores a computer program or instruction, which, when executed by a processor, implements the steps of the image detection method according to any one of claims 1 to 10.

14. A computer program product, characterized in that The method comprises a computer program or instructions, which, when executed by a processor, implements the steps of the image detection method according to any one of claims 1 to 10.

Images