Fuzzy image recognition method, device, equipment and storage medium

Abstract

The present invention relates to the field of image processing technology, and discloses a fuzzy image recognition method, device, equipment and storage medium. The present invention obtains a target image to be recognized by preprocessing the image to be recognized, then obtains the parameters to be recognized corresponding to the target image to be recognized, and then performs fuzzy image recognition on the image to be recognized according to preset test parameters and the parameters to be recognized. The present invention obtains the parameters to be recognized corresponding to the target image to be recognized, and then performs fuzzy image recognition on the image to be recognized according to the preset test parameters and the parameters to be recognized. Compared with the existing method of determining the blur degree of an image by the second-order derivative response of Laplace, which is only applicable to the recognition of an overall fuzzy image, the above-mentioned method of the present invention performs fuzzy image recognition according to the preset test parameters and the parameters to be recognized corresponding to the image to be recognized, can accurately detect a local fuzzy image, thereby accurately recognizing a fuzzy image, and can improve the fuzzy image recognition speed.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a fuzzy image recognition method, device, equipment and storage medium. Background Art

[0002] In the OLED panel manufacturing process, AOI (Automated Optical Inspection) equipment automatically detects defects in panel images by combining optical imaging systems with related image processing technologies. However, due to image parameter adjustments in the AOI system or defocus issues caused by product protrusion anomalies, a large number of blurred images are inevitably generated. To further optimize the optical imaging system parameters and detect product anomalies, it is necessary to identify and recall these blurred images and calculate their percentage. Therefore, accurately identifying blurred images has become a pressing issue.

[0003] The above content is only used to assist in understanding the technical solution of the present invention and does not constitute an admission that the above content is prior art. Summary of the Invention

[0004] The main purpose of the present invention is to provide a blurred image recognition method, device, equipment and storage medium, aiming to solve the technical problem that the existing technology cannot accurately recognize blurred images.

[0005] To achieve the above object, the present invention provides a fuzzy image recognition method, which includes:

[0006] Preprocess the image to be identified to obtain the target image to be identified;

[0007] Obtaining parameters to be identified corresponding to the target image to be identified;

[0008] Fuzzy image recognition is performed on the image to be recognized according to preset test parameters and the parameters to be recognized.

[0009] Optionally, the step of preprocessing the image to be identified to obtain a target image to be identified specifically includes:

[0010] Grayscale processing is performed on the image to be identified to obtain the processed image to be identified;

[0011] Performing grid segmentation on the processed image to be identified to obtain a segmented image to be identified;

[0012] De-noising is performed on the segmented image to be identified to obtain a target image to be identified.

[0013] Optionally, the step of obtaining parameters to be identified corresponding to the target image to be identified specifically includes:

[0014] Obtaining a significant edge response to be identified corresponding to the target image to be identified according to a preset edge detection operator;

[0015] Determining a variance set to be identified according to the significant edge response to be identified;

[0016] Sorting all variances in the variance set to be identified to obtain an ordered variance set to be identified;

[0017] The to-be-identified parameters corresponding to the target to-be-identified image are determined according to the ordered to-be-identified variance set.

[0018] Optionally, the parameters to be identified include: fuzzy parameters to be identified and non-fuzzy parameters to be identified;

[0019] The step of determining the parameters to be identified corresponding to the target image to be identified according to the ordered variance set to be identified specifically includes:

[0020] Performing mean processing on the variances of a first preset order in the ordered set of variances to be identified, to obtain a fuzzy parameter to be identified corresponding to the target image to be identified;

[0021] The mean of all variances in the ordered variance set to be identified is obtained, and the unambiguous parameter to be identified corresponding to the target image to be identified is determined according to the mean and variances of a second preset order in the ordered variance set to be identified.

[0022] Optionally, before the step of preprocessing the image to be identified to obtain the target image to be identified, the step further includes:

[0023] Acquire a test image of a preset type, and preprocess the test image of the preset type to obtain a target test image of the preset type;

[0024] Obtaining a variance set corresponding to the target test image of the preset type;

[0025] The preset test parameters corresponding to the target test image are determined according to the variance set.

[0026] Optionally, the preset test parameters include: test fuzzy parameters and test non-fuzzy parameters;

[0027] The step of determining the preset test parameters corresponding to the target test image according to the variance set specifically includes:

[0028] Sorting the variances in the variance sets of each preset type respectively to obtain the ordered variance sets of each preset type;

[0029] performing mean processing on the variances of a third preset order in the sorted variance sets of each preset type to obtain a test blur parameter corresponding to the target test image;

[0030] Obtain a mean of all variances in the sorted variance sets of each preset type, and determine a test unblurring parameter corresponding to the target test image according to the mean and variances of a fourth preset order in the sorted variance sets of each preset type.

[0031] Optionally, the step of performing fuzzy image recognition on the image to be recognized according to preset test parameters and the parameters to be recognized specifically includes:

[0032] determining a first threshold value according to the test fuzzy parameter;

[0033] determining a second threshold value based on the test non-ambiguous parameter;

[0034] When the fuzzy parameter to be identified is smaller than the first threshold and the non-fuzzy parameter to be identified is smaller than a second threshold, it is determined that the image to be identified is a fuzzy image.

[0035] In addition, to achieve the above-mentioned object, the present invention further provides a fuzzy image recognition device, the fuzzy image recognition device comprising:

[0036] An image processing module is used to pre-process the image to be identified to obtain a target image to be identified;

[0037] A parameter acquisition module, used to acquire the parameters to be identified corresponding to the target image to be identified;

[0038] The fuzzy image recognition module is used to perform fuzzy image recognition on the image to be recognized according to preset test parameters and the parameters to be recognized.

[0039] In addition, to achieve the above-mentioned purpose, the present invention also proposes a blurred image recognition device, which includes: a memory, a processor, and a blurred image recognition program stored in the memory and runnable on the processor, and the blurred image recognition program is configured to implement the blurred image recognition method described above.

[0040] In addition, to achieve the above-mentioned purpose, the present invention further proposes a storage medium, on which a fuzzy image recognition program is stored. When the fuzzy image recognition program is executed by a processor, the fuzzy image recognition method described above is implemented.

[0041] The present invention preprocesses an image to be identified to obtain a target image to be identified, then obtains parameters to be identified corresponding to the target image to be identified, and then performs fuzzy image recognition on the image to be identified based on preset test parameters and the parameters to be identified. Compared to existing methods that determine the blur level of an image using the Laplace second-order derivative response, which are only applicable to the recognition of globally blurred images, the above-mentioned method of the present invention performs fuzzy image recognition based on preset test parameters and the parameters to be identified corresponding to the image to be identified, accurately detecting locally blurred images, thereby accurately identifying blurred images, and improving the speed of blurred image recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a structural diagram of a fuzzy image recognition device in a hardware operating environment involved in an embodiment of the present invention;

[0043] Figure 2 Schematic diagram of the flow of the first embodiment of the fuzzy image recognition method of the present invention;

[0044] Figure 3 Schematic diagram of the flow of the second embodiment of the fuzzy image recognition method of the present invention;

[0045] Figure 4 Schematic diagram of the flow of the third embodiment of the fuzzy image recognition method of the present invention;

[0046] Figure 5 FIG. 4 is a structural block diagram of the first embodiment of the fuzzy image recognition device of the present invention.

[0047] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0048] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0049] Reference Figure 1 , Figure 1 This is a schematic diagram of the structure of a fuzzy image recognition device in the hardware operating environment involved in an embodiment of the present invention.

[0050] like Figure 1As shown, the blurred image recognition device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the user interface 1003 may optionally include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM) or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0051] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the blurred image recognition device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0052] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a fuzzy image recognition program.

[0053] exist Figure 1 In the fuzzy image recognition device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the fuzzy image recognition device of the present invention can be set in the fuzzy image recognition device, and the fuzzy image recognition device calls the fuzzy image recognition program stored in the memory 1005 through the processor 1001 and executes the fuzzy image recognition method provided by the embodiment of the present invention.

[0054] The embodiment of the present invention provides a fuzzy image recognition method, referring to Figure 2 , Figure 2 FIG. 4 is a flow chart of the first embodiment of the fuzzy image recognition method of the present invention.

[0055] In this embodiment, the blurred image recognition method includes the following steps:

[0056] Step S10: pre-processing the image to be identified to obtain a target image to be identified;

[0057] It should be noted that the execution subject of this embodiment can be the above-mentioned elevator floor error judgment device with image processing, network communication and program running functions, or other devices that can achieve the same or similar functions. This embodiment does not impose specific restrictions on this.

[0058] It is understandable that the image to be identified refers to an image captured by an optical imaging system in the production process of an organic light-emitting diode (OLED) panel.

[0059] Furthermore, in order to accurately obtain the target image to be identified, in this embodiment, step S10 includes: grayscale processing of the image to be identified to obtain a processed image to be identified; grid-segmenting the processed image to be identified to obtain a segmented image to be identified; and denoising the segmented image to be identified to obtain the target image to be identified.

[0060] It can be understood that grayscaling refers to processing the RGB color components of the image to be identified so that all RGB color components of the image to be identified are equal, and the corresponding values ​​are grayscale values. According to the YUV color space, the physical meaning of the Y component is the brightness of the point, and this value reflects the brightness level. Based on the changing relationship between the RGB and YUV color spaces, the correspondence between the brightness Y and the three color components R, G, and B can be established: Y = 0.3R + 0.59G + 0.11B. This brightness value is used to express the grayscale value of the processed image to be identified.

[0061] It should be understood that grid segmentation refers to the uniform division of the processed image to be identified in the horizontal and vertical directions to obtain a fixed number of local image blocks, where the horizontal and vertical numbers can be set according to the actual image situation, and this embodiment does not impose specific restrictions on this. Specifically, in the grid segmentation process, in order to ensure that image blur recognition is not affected by the image boundary, the outer boundaries of the 4 pixels above, below, left and right of the image can be removed, and the remaining center part is used for grid division. At the same time, in order to ensure that the size of the divided grid is an integer pixel, the redundant boundary pixels of the image can be removed by calculating the remainder of the division of the image size and the number of grid divisions, and then used for grid division.

[0062] In practice, this embodiment uses a Gaussian kernel to smooth and denoise the processed image. Gaussian kernel denoising involves performing a weighted average of the entire image. The value of each pixel is the weighted average of its own value and the values ​​of other pixels in its neighborhood. Gaussian kernel denoising involves scanning each pixel in the image using a template (also known as a convolution or mask) and replacing the value of the central pixel in the template with the weighted average grayscale value of the pixels in the neighborhood defined by the template.

[0063] This embodiment can divide a large-sized image into smaller image blocks by grayscale processing, grid segmentation and denoising processing on the image to be identified, and can realize parallel processing of a large number of image blocks in batches, thereby greatly reducing the blur detection time of large-sized images and reducing local noise interference in the obtained target image to be identified.

[0064] Step S20: Obtaining parameters to be identified corresponding to the target image to be identified;

[0065] It can be understood that the parameters to be identified refer to the parameters corresponding to the target image to be identified, which may specifically include blur parameters to be identified and non-blur parameters to be identified. The blur parameters to be identified are used to indicate the overall average blur degree of the target image to be identified, and the non-blur parameters to be identified are used to distinguish non-blurred images with a small amount of significant texture.

[0066] Step S30: performing fuzzy image recognition on the image to be recognized according to preset test parameters and the parameters to be recognized.

[0067] It should be understood that the preset test parameters also include test module parameters and test non-fuzzy parameters, so that the image to be identified can be fuzzy image recognized through the test module parameters, the test non-fuzzy parameters, the fuzzy parameters to be identified, and the size relationship between the non-fuzzy parameters to be identified to determine whether the image to be identified is a fuzzy image.

[0068] This embodiment preprocesses the image to be identified to obtain a target image to be identified, then obtains parameters to be identified corresponding to the target image to be identified, and then performs blurred image recognition on the image to be identified based on preset test parameters and the parameters to be identified. This embodiment obtains parameters to be identified corresponding to the target image to be identified, and then performs blurred image recognition on the image to be identified based on the preset test parameters and the parameters to be identified. Compared to existing methods that determine the blur level of an image using the second-order derivative response of Laplace, which are only applicable to the recognition of globally blurred images, this embodiment performs blurred image recognition based on preset test parameters and the parameters to be identified corresponding to the image to be identified, accurately detecting locally blurred images, thereby accurately identifying blurred images and improving the speed of blurred image recognition.

[0069] refer to Figure 3 , Figure 3 FIG. 4 is a flow chart of a second embodiment of a fuzzy image recognition method according to the present invention.

[0070] Based on the above first embodiment, in this embodiment, step S20 includes:

[0071] Step S201: obtaining a significant edge response to be identified corresponding to the target image to be identified according to a preset edge detection operator;

[0072] It should be noted that the preset edge detection operator refers to a pre-set edge detection operator, such as a Canny operator, a Sobel operator, a Laplace operator, etc., and this embodiment does not impose any specific limitation on this.

[0073] In a specific implementation, edge detection can be performed on the target image to be identified using a preset edge detection operator to obtain the corresponding salient edge response to be identified for the target image to be identified. In this embodiment, since the target image to be identified is obtained after gridding and segmentation, the target image to be identified includes multiple image blocks. Therefore, it is necessary to calculate the Laplace second-order response for all image blocks in the target image to obtain the salient edge response corresponding to each image block.

[0074] Step S202: determining a variance set to be identified according to the significant edge response to be identified;

[0075] It should be noted that the variance to be identified can be used to indicate the smoothness of each local image block in the target image to be identified. The smaller the variance to be identified, the more blurred the target image to be identified.

[0076] It can be understood that the variance set to be identified includes the variances corresponding to all the significant edge responses to be identified. There are multiple significant edge responses to be identified, and there are also multiple corresponding variances to be identified. The variances to be identified corresponding to all image blocks in the target image to be identified can be determined through each significant edge response to be identified.

[0077] Step S203: sorting all variances in the variance set to be identified to obtain an ordered variance set to be identified;

[0078] In a specific implementation, the sorting of all variances in this embodiment may be from small to large or from large to small, and this embodiment does not impose any specific limitation on this.

[0079] Step S204: determining the to-be-identified parameters corresponding to the target to-be-identified image according to the ordered to-be-identified variance set.

[0080] It is understandable that the parameters to be identified in this embodiment include fuzzy parameters to be identified and non-fuzzy parameters to be identified, and specific parameter values ​​can be calculated based on all variances in the ordered variance set to be identified.

[0081] Furthermore, in order to accurately determine the parameters to be identified, in this embodiment, step S204 includes: performing mean processing on the variances of the first preset order in the ordered variance set to be identified to obtain the fuzzy parameters to be identified corresponding to the target image to be identified; obtaining the mean of all variances in the ordered variance set to be identified, and determining the non-fuzzy parameters to be identified corresponding to the target image to be identified based on the mean and the variances of the second preset order in the ordered variance set to be identified.

[0082] It should be noted that the first preset order and the second preset order are pre-set orders in the ordered set of variances to be identified, for example: the first to 30th variances, the first to 50th variances, etc., and this embodiment does not impose any specific restrictions on this.

[0083] It is understandable that if the number of image blocks of the target image to be identified after gridding and segmentation is 77, when the variances in the ordered variance set to be identified are arranged in order from small to large, the first preset order can be set to the first 50, that is, the average of the first 50 variances is calculated, and the average is the fuzzy parameter to be identified corresponding to the target image to be identified; when the variances in the ordered variance set to be identified are arranged in order from large to small, the first preset order can be set to the last 50, that is, the average of the last 50 variances is calculated, and the average is the fuzzy parameter to be identified corresponding to the target image to be identified.

[0084] In a specific implementation, it is necessary to calculate the mean of all variances in the ordered set of variances to be identified. When the variances in the ordered set of variances to be identified are arranged in order from small to large, the second preset order can be set to the last 10, and the mean square of the difference between the last 10 variances and the mean of all variances is the non-fuzzy parameter to be identified corresponding to the target image to be identified; when the variances in the ordered set of variances to be identified are arranged in order from large to small, the second preset order can be set to the first 10, and the mean square of the difference between the first 10 variances and the mean of all variances is the non-fuzzy parameter to be identified corresponding to the target image to be identified.

[0085] This embodiment obtains salient edge responses corresponding to the target image to be identified based on a preset edge detection operator, then determines a variance set to be identified based on the salient edge responses. All variances in the variance set to be identified are then sorted to obtain an ordered variance set to be identified, and parameters to be identified corresponding to the target image to be identified are determined based on the ordered variance set to be identified. This embodiment, by sorting all variances in the variance set to obtain an ordered variance set to be identified, and then determining parameters to be identified corresponding to the target image to be identified based on the ordered variance set to be identified, can detect local salient edges within a grid and, combined with the local maximum variance, obtain accurate parameters to be identified.

[0086] refer to Figure 4 , Figure 4 FIG. 4 is a flow chart of a third embodiment of a fuzzy image recognition method according to the present invention.

[0087] Based on the above embodiments, in this embodiment, before step S10, the method further includes:

[0088] Step S00: obtaining a test image of a preset type, and preprocessing the test image of the preset type to obtain a target test image of the preset type;

[0089] It is understandable that the preset types may include a solid color background type, a clear type, and a blurred type, and the number of different types of test images is also different. In this embodiment, 100 test images of a solid color background type, 300 test images of a clear type, and 600 test images of a blurred type can be selected. The specific number can be set according to actual conditions, and this embodiment does not impose any specific restrictions on this.

[0090] It should be understood that the preset type of test image is preprocessed, that is, the preset type of test image is grayscaled, gridded and denoised. The specific process can refer to the above-mentioned steps of preprocessing the image to be identified, and this embodiment will not go into details about this.

[0091] Step S01: obtaining a variance set corresponding to a target test image of a preset type;

[0092] It is understandable that target test images of different preset types correspond to different variance sets, and the number of variances in the variance set is the number of image blocks in the target test image.

[0093] Step S02: determining preset test parameters corresponding to the target test image according to the variance set.

[0094] It is understandable that the preset test parameters in this embodiment include test fuzzy parameters and test non-fuzzy parameters, and the specific parameter values ​​can be calculated based on all variances in the variance set.

[0095] Furthermore, in this embodiment, the step S02 includes: sorting the variances in the variance sets of each preset type respectively to obtain the ordered variance sets of each preset type; performing mean processing on the variances of the third preset order in the sorted variance sets of each preset type to obtain the test blur parameters corresponding to the target test image; obtaining the mean of all variances in the sorted variance sets of each preset type, and determining the test non-blur parameters corresponding to the target test image based on the mean and the variances of the fourth preset order in the sorted variance sets of each preset type.

[0096] It should be noted that this embodiment sorts all variances in each preset variance set, either from smallest to largest or from largest to smallest, and this embodiment does not impose any specific restrictions on this. The third and fourth preset orders are pre-set orders within the ordered variance set, for example, the first to 30th variances, the first to 50th variances, etc., and this embodiment does not impose any specific restrictions on this.

[0097] It is understandable that after sorting all variances in the variance sets of each preset type respectively, an ordered variance set of a pure color background type, an ordered variance set of a clear type, and an ordered variance set of a fuzzy type can be obtained.

[0098] It should be understood that if the target test image has 77 image blocks after gridding, the test fuzzy parameter corresponding to the target test image of the solid background type is the first fuzzy parameter, the test fuzzy parameter corresponding to the target test image of the clear type is the second fuzzy parameter, and the test fuzzy parameter corresponding to the target test image of the fuzzy type is the third fuzzy parameter. When the variances in the ordered variance set are arranged in order from small to large, the third preset order can be set to the first 50, that is, the first 50 variances in the variance set of each preset type are averaged to obtain the first fuzzy parameter, the second fuzzy parameter, and the third fuzzy parameter; when the variances in the ordered variance set to be identified are arranged in order from large to small, the first preset order can be set to the last 50, that is, the last 50 variances in the variance set of each preset type are averaged to obtain the first fuzzy parameter, the second fuzzy parameter, and the third fuzzy parameter.

[0099] In a specific implementation, it is assumed that the test non-fuzzy parameter corresponding to the target test image of the solid color background type is the first non-fuzzy parameter, the test non-fuzzy parameter corresponding to the target test image of the clear type is the second non-fuzzy parameter, and the test non-fuzzy parameter corresponding to the target test image of the blurry type is the third non-fuzzy parameter. It is necessary to calculate the mean of all variances in the ordered variance set of each preset type. When the variances in the ordered variance set are arranged in order from small to large, the fourth preset order can be set to the last 10, and the mean square of the difference between the last 10 variances and the mean of all variances is the test non-fuzzy parameter corresponding to the target test image of each preset type; when the variances in the ordered variance set are arranged in order from large to small, the fourth preset order can be set to the first 10, and the mean square of the difference between the first 10 variances and the mean of all variances is the test non-fuzzy parameter corresponding to the target test image of each preset type.

[0100] Furthermore, in this embodiment, the step S30 includes: determining a first threshold value based on the test fuzzy parameter; determining a second threshold value based on the test non-fuzzy parameter; and determining that the image to be identified is a blurred image when the fuzzy parameter to be identified is less than the first threshold value and the non-fuzzy parameter to be identified is less than the second threshold value.

[0101] It can be understood that the first threshold can be obtained based on the second blur parameter and the third blur parameter. For example, the second blur parameter values ​​of all clear test images are distributed between 1.385 and 14.341, and the third blur parameter values ​​of all blurry test images are distributed between 0.064 and 0.637. Therefore, the first threshold value can be set between 0.637 and 1.385, and an intermediate value of 1.0 can be taken as the first threshold value.

[0102] It should be understood that the second threshold can be obtained based on the first non-fuzzy parameter and the third non-fuzzy parameter. For example, the first non-fuzzy parameter values ​​of all test images with solid background types are distributed between 0.116 and 37.352, and the third non-fuzzy parameter values ​​of all test images with fuzzy types are distributed between 0.00015 and 0.274. The two distribution ranges have an intersection, and the number of fuzzy images between 0.116 and 0.274 is 17, accounting for 2.8% of the total fuzzy images. Therefore, the second threshold can be set to 0.116, so that only 2.8% of the fuzzy images will be missed.

[0103] In a specific implementation, when the fuzzy parameter to be identified corresponding to the image to be identified is less than a first threshold and the non-fuzzy parameter to be identified corresponding to the image to be identified is less than a second threshold, the image to be identified is determined to be a fuzzy image; when the fuzzy parameter to be identified corresponding to the image to be identified is less than the first threshold and the non-fuzzy parameter to be identified corresponding to the image to be identified is greater than or equal to the second threshold, or, the fuzzy parameter to be identified corresponding to the image to be identified is greater than or equal to the first threshold and the non-fuzzy parameter to be identified corresponding to the image to be identified is less than the second threshold, or, the fuzzy parameter to be identified corresponding to the image to be identified is greater than or equal to the first threshold and the non-fuzzy parameter to be identified corresponding to the image to be identified is greater than or equal to the second threshold, the image to be identified is determined to be a non-fuzzy image.

[0104] This embodiment obtains a preset test image type and preprocesses it to obtain a target test image of the preset type. A variance set corresponding to the target test image of the preset type is then obtained, and then preset test parameters corresponding to the target test image are determined based on the variance set. By obtaining multiple test images of each preset type, this embodiment can more accurately obtain the preset test parameters, thereby making the blurred image recognition process based on the preset test parameters and the parameters corresponding to the image to be recognized more accurate, enabling accurate recognition of blurred images.

[0105] In addition, an embodiment of the present invention further provides a storage medium, on which a fuzzy image recognition program is stored. When the fuzzy image recognition program is executed by a processor, the fuzzy image recognition method described above is implemented.

[0106] Reference Figure 5 , Figure 5 FIG. 4 is a structural block diagram of the first embodiment of the fuzzy image recognition device of the present invention.

[0107] like Figure 5 As shown, the fuzzy image recognition device proposed in the embodiment of the present invention includes:

[0108] The image processing module 10 is used to pre-process the image to be identified to obtain the target image to be identified;

[0109] The parameter acquisition module 20 is used to obtain the parameters to be identified corresponding to the target image to be identified;

[0110] The fuzzy image recognition module 30 is configured to perform fuzzy image recognition on the image to be recognized according to preset test parameters and the parameters to be recognized.

[0111] This embodiment preprocesses the image to be identified to obtain a target image to be identified, then obtains parameters to be identified corresponding to the target image to be identified, and then performs blurred image recognition on the image to be identified based on preset test parameters and the parameters to be identified. This embodiment obtains parameters to be identified corresponding to the target image to be identified, and then performs blurred image recognition on the image to be identified based on the preset test parameters and the parameters to be identified. Compared to existing methods that determine the blur level of an image using the second-order derivative response of Laplace, which are only applicable to the recognition of globally blurred images, this embodiment performs blurred image recognition based on preset test parameters and the parameters to be identified corresponding to the image to be identified, accurately detecting locally blurred images, thereby accurately identifying blurred images and improving the speed of blurred image recognition.

[0112] Based on the first embodiment of the blurred image recognition device of the present invention, a second embodiment of the blurred image recognition device of the present invention is proposed.

[0113] In this embodiment, the image processing module 10 is further used to perform grayscale processing on the image to be identified to obtain a processed image to be identified; perform grid segmentation on the processed image to be identified to obtain a segmented image to be identified; and perform denoising on the segmented image to be identified to obtain a target image to be identified.

[0114] Furthermore, the parameter acquisition module 20 is also used to obtain the significant edge response to be identified corresponding to the target image to be identified according to a preset edge detection operator; determine the variance set to be identified according to the significant edge response to be identified; sort all variances in the variance set to be identified to obtain an ordered variance set to be identified; and determine the parameters to be identified corresponding to the target image to be identified according to the ordered variance set to be identified.

[0115] Furthermore, the parameter acquisition module 20 is also used to perform mean processing on the variances of the first preset order in the ordered variance set to be identified to obtain the fuzzy parameters to be identified corresponding to the target image to be identified; obtain the mean of all variances in the ordered variance set to be identified, and determine the non-fuzzy parameters to be identified corresponding to the target image to be identified based on the mean and the variances of the second preset order in the ordered variance set to be identified.

[0116] Furthermore, the fuzzy image recognition module 30 is also used to obtain a preset type of test image, and preprocess the preset type of test image to obtain a preset type of target test image; obtain a variance set corresponding to the preset type of target test image; and determine the preset test parameters corresponding to the target test image based on the variance set.

[0117] Furthermore, the blurred image recognition module 30 is also used to sort the variances in the variance sets of each preset type respectively to obtain the ordered variance sets of each preset type; perform mean processing on the variances of the third preset order in the sorted variance sets of each preset type to obtain the test blur parameters corresponding to the target test image; obtain the mean of all variances in the sorted variance sets of each preset type, and determine the test non-blurred parameters corresponding to the target test image based on the mean and the variances of the fourth preset order in the sorted variance sets of each preset type.

[0118] Furthermore, the blurred image recognition module 30 is also used to determine a first threshold value based on the test blurred parameter; determine a second threshold value based on the test non-fuzzy parameter; and determine that the image to be recognized is a blurred image when the blurred parameter to be recognized is less than the first threshold value and the non-fuzzy parameter to be recognized is less than the second threshold value.

[0119] Other embodiments or specific implementations of the fuzzy image recognition device of the present invention can refer to the above-mentioned method embodiments and will not be described in detail here.

[0120] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0121] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0122] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0123] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A fuzzy image recognition method, characterized in that: The fuzzy image recognition method comprises: Acquire a test image of a preset type, and preprocess the test image of the preset type to obtain a target test image of a preset type, wherein the preset types include: a solid color background type, a clear type, and a fuzzy type; Obtaining a variance set corresponding to the target test image of the preset type; Sorting the variances in the variance sets of each preset type respectively to obtain the ordered variance sets of each preset type; performing mean processing on the variances of a third preset order in the sorted variance sets of each preset type to obtain a test blur parameter corresponding to the target test image; Obtaining a mean of all variances in the sorted variance sets of each preset type, and determining a test unambiguous parameter corresponding to the target test image according to the mean and variances of a fourth preset order in the sorted variance sets of each preset type; Preprocess the image to be identified to obtain the target image to be identified; Obtaining a significant edge response to be identified corresponding to the target image to be identified according to a preset edge detection operator; Determining a variance set to be identified according to the significant edge response to be identified; Sorting all variances in the variance set to be identified to obtain an ordered variance set to be identified; Performing mean processing on the variances of a first preset order in the ordered set of variances to be identified, to obtain a fuzzy parameter to be identified corresponding to the target image to be identified; Obtaining a mean of all variances in the ordered set of variances to be identified, and determining a non-blurred parameter to be identified corresponding to the target image to be identified based on the mean and variances of a second preset order in the ordered set of variances to be identified, wherein the fuzzy parameter to be identified is used to indicate an overall average blur degree of the target image to be identified, and the non-blurred parameter to be identified is used to distinguish non-blurred images; determining a first threshold value according to the test fuzzy parameter; determining a second threshold value based on the test non-ambiguous parameter; When the fuzzy parameter to be identified is smaller than the first threshold and the non-fuzzy parameter to be identified is smaller than a second threshold, it is determined that the image to be identified is a fuzzy image.

2. The fuzzy image recognition method according to claim 1, wherein: The step of preprocessing the image to be identified to obtain the target image to be identified specifically includes: Grayscale processing is performed on the image to be identified to obtain the processed image to be identified; Performing grid segmentation on the processed image to be identified to obtain a segmented image to be identified; De-noising is performed on the segmented image to be identified to obtain a target image to be identified.

3. A fuzzy image recognition device, characterized in that: The fuzzy image recognition device comprises: An image processing module is used to pre-process the image to be identified to obtain a target image to be identified; A parameter acquisition module is configured to acquire, based on a preset edge detection operator, a significant edge response to be identified corresponding to the target image to be identified; determine a variance set to be identified based on the significant edge response to be identified; sort all variances in the variance set to be identified to obtain an ordered variance set to be identified; perform mean processing on variances of a first preset order in the ordered variance set to be identified to obtain a fuzzy parameter to be identified corresponding to the target image to be identified; acquire the mean of all variances in the ordered variance set to be identified, and determine a non-fuzzy parameter to be identified corresponding to the target image to be identified based on the mean and variances of a second preset order in the ordered variance set to be identified, wherein the fuzzy parameter to be identified is used to indicate an overall average degree of blur of the target image to be identified, and the non-fuzzy parameter to be identified is used to distinguish a non-fuzzy image; A fuzzy image recognition module is configured to obtain a test image of a preset type, pre-process the test image of the preset type, and obtain a target test image of the preset type; obtain a variance set corresponding to the target test image of the preset type; sort the variances in the variance set of each preset type respectively to obtain an ordered variance set of each preset type; perform mean processing on the variances of a third preset order in the sorted variance set of each preset type to obtain a test fuzzy parameter corresponding to the target test image; obtain the mean of all variances in the sorted variance set of each preset type, and determine the test non-fuzzy parameter corresponding to the target test image based on the mean and the variances of a fourth preset order in the sorted variance set of each preset type, wherein the preset types include: a solid color background type, a clear type, and a fuzzy type; The blurred image recognition module is further used to determine a first threshold based on the test blurred parameter; determine a second threshold based on the test non-fuzzy parameter; and determine that the image to be recognized is a blurred image when the blurred parameter to be recognized is less than the first threshold and the non-fuzzy parameter to be recognized is less than the second threshold.

4. A fuzzy image recognition device, characterized in that: The blurred image recognition device includes: a memory, a processor, and a blurred image recognition program stored in the memory and executable on the processor, wherein the blurred image recognition program is configured to implement the blurred image recognition method according to claim 1 or 2.

5. A storage medium, characterized in that: The storage medium stores a fuzzy image recognition program, and when the fuzzy image recognition program is executed by the processor, the fuzzy image recognition method according to claim 1 or 2 is implemented.

Images