An image feature enhancement method and device for monitoring equipment

Abstract

The application discloses an image feature enhancement method and device for monitoring equipment, and relates to the field of digital image processing; an original image is acquired, the original image is preprocessed to obtain a brightness mean value, and the original image is averagely divided to obtain a plurality of image regions; image features of each image region are extracted, and an image mark is added to the image region according to the image features; the image regions are stored in corresponding mark libraries according to the image marks; the image regions in a detail mark library and a noise point mark library are respectively enhanced; the enhanced image regions in the coordinate mark library, the detail mark library and the noise point mark library are spliced according to coordinates to obtain a fusion image, and the brightness of pixel points in the fusion image is adjusted according to the brightness mean value; the processing method can more effectively perform fine processing on different parts of the image, improve the overall quality of the image, and meanwhile, the accuracy of color and the authenticity of details are maintained.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of digital image processing, and particularly relates to an image feature enhancement method and device for monitoring equipment. BACKGROUND

[0002] A monitoring system is a comprehensive system integrating video capture, transmission, storage and display technologies, which is used for real-time monitoring and recording of specific areas. It can provide real-time video images to help monitoring personnel discover abnormal situations in time and take appropriate measures, and is widely used in public security, traffic management, enterprise security and other fields.

[0003] In a weak light environment, images are often characterized by low brightness, low contrast and high noise due to insufficient light, which makes it difficult to identify details and key information in the image. In order to improve image quality, features such as color, detail and noise in the image are analyzed to enhance the distinguishability of the image. In a weak light environment, color information may be distorted, so color correction is used to restore the true color distribution. At the same time, contrast and brightness enhancement algorithms are used to improve the contrast and brightness of the image, so that the details in the dark part can be revealed. In addition, detail enhancement techniques are also applied to highlight the detailed features in the image and enhance the clarity of the image. The application of these technologies can improve the distinguishability of images taken in a weak light environment and provide some reference value for subsequent image analysis.

[0004] The above-mentioned technology can improve the image quality in a weak light environment, but also has disadvantages. For example, color correction and enhancement algorithms may not completely restore the true color information, resulting in distorted image colors. In addition, contrast and brightness enhancement may introduce noise or over-enhancement, affecting the naturalness and authenticity of the image. Detail enhancement may also cause the image to be over-sharpened and lose details. Excessive uniform optimization makes the image not useful. SUMMARY

[0005] The purpose of the present application is to solve the problem of excessive uniform optimization making the image not useful, and to propose an image feature enhancement method and device for monitoring equipment.

[0006] In the first aspect of the present application, an image feature enhancement method for monitoring equipment is first proposed, which comprises:

[0007] Optionally, the method comprises:

[0008] An original image is obtained, the original image is preprocessed to obtain a brightness mean value, and the original image is divided into a plurality of image regions;

[0009] Converting the image region from a spatial domain to a frequency domain, calculating a mean value of a high frequency part in the frequency domain to obtain a definition, and adding a detail mark to the image region if the definition exceeds a preset detail value;

[0010] Obtaining a pixel point in the image region as a noise point if a value of the pixel point deviates from values of surrounding pixel points by more than a threshold value, and adding a noise point mark to the image region if a number of the noise points exceeds a preset number.

[0011] Enhancing the image region in the detail mark library and the noise point mark library respectively to obtain an enhanced image;

[0012] Splicing the enhanced image region in the coordinate mark library, the detail mark library and the noise point mark library according to coordinates to obtain a fusion image, and adjusting brightness of a pixel point in the fusion image according to a brightness mean value.

[0013] Optionally, obtaining an original image, and pre-processing the original image to obtain the brightness mean value includes:

[0014] Converting the original image from an RGB color space to an XYZ color space to obtain first data, converting the first data from the XYZ color space to a Lab color space to obtain second data, and binarizing the second data to obtain third data;

[0015] Calculating an average brightness of the image in the third data as the brightness mean value.

[0016] Optionally, adding an image mark to the image region according to an image feature includes:

[0017] Converting the image region from a spatial domain to a frequency domain using a frequency domain analysis method to calculate a mean value of a high frequency part in the frequency domain to obtain a definition, and adding a detail mark to the image region if the definition exceeds a preset detail value;

[0018] Obtaining a pixel point in the image region as a noise point if a value of the pixel point deviates from values of surrounding pixel points by more than a threshold value, and adding a noise point mark to the image region if a number of the noise points exceeds a preset number.

[0019] Optionally, enhancing the image region in the detail mark library and the noise point mark library respectively to obtain an enhanced image includes:

[0020] If the detail mark library has a detail mark, performing a detail enhancement processing on the image region in the region mark library having the detail mark by histogram equalization to obtain an enhanced region;

[0021] If the noise point mark library has a noise point mark, performing a noise reduction processing on the image region in the region mark library having the noise point mark by Gaussian filtering to obtain an enhanced region.

[0022] Optionally, adjusting brightness of a pixel point in the fusion image according to a brightness mean value includes:

[0023] Converting the fusion image from the RGB color space to the Lab color space and calculating correction parameters of the a channel and the b channel at the pixel point (i, j) through a color correction formula With By With Correcting the color of the pixel point (i, j) in the a channel and the b channel to obtain a restoration image;

[0024] Adjusting any pixel point in the restoration image through the brightness mean value and inversely converting the restoration image from the Lab color space to the RGB color space to obtain a final enhanced image;

[0025] The color correction formula is:

[0026]

[0027] Wherein, M*N is the image size, D a (i,j) and D b (i,j) are the pixels at (i, j) in the a channel and the b channel respectively.

[0028] In the second aspect of the embodiment of the present application, an image feature enhancement device for monitoring equipment is provided, comprising:

[0029] Optionally, the device comprises a preprocessing module, a mark adding module, a mark storage module, an image enhancement module and an image fusion module:

[0030] The preprocessing module is used to acquire an original image, pre-process the original image to obtain a brightness mean value, and average-divide the original image to obtain a plurality of image regions;

[0031] The mark adding module is used to extract image features of each image region, and add image marks to the image region according to the image features; the image features include the definition of image details and the number of noise points; the image marks include coordinates, detail marks and noise point marks; if the definition value of the image region exceeds a preset detail value, a detail mark is added; if the noise points of the image region exceed a threshold value, a noise point mark is added;

[0032] The mark storage module is used to store each image region in a corresponding mark library according to the image marks; the mark library includes a coordinate mark library, a detail mark library and a noise point mark library; the image regions in the coordinate mark library only contain coordinates, the image regions in the detail mark library contain detail marks, and the image regions in the noise point mark library contain noise point marks;

[0033] The image enhancement module is used to enhance the image regions in the detail mark library and the noise point mark library respectively.

[0034] The image fusion module is configured to splice the enhanced image regions in the coordinate mark library, the detail mark library and the noise mark library according to the coordinates to obtain a fusion image, and adjust the brightness of the pixel points in the fusion image according to the brightness mean value.

[0035] Optionally, the preprocessing module comprises an image conversion module and a brightness calculation module.

[0036] The image conversion module is configured to convert the original image from the RGB color space to the XYZ color space to obtain first data, convert the first data from the XYZ color space to the Lab color space to obtain second data, and perform binaryzation processing on the second data to obtain third data.

[0037] The brightness calculation module is configured to calculate the average brightness of the image in the third data as the brightness mean value.

[0038] Optionally, the mark adding module comprises a detail mark module and a noise mark module.

[0039] The detail mark module is configured to convert the image region from the spatial domain to the frequency domain, calculate the mean value of the high-frequency part in the frequency domain to obtain the definition, and add a detail mark to the image region if the definition exceeds a preset detail value.

[0040] The noise mark module is configured to obtain the pixel points in the image region that deviate from the numerical values of the surrounding pixel points by more than a threshold value as noise points, and add a noise mark to the image region if the number of the noise points exceeds a preset number.

[0041] Optionally, the image enhancement module comprises a detail enhancement module and a noise reduction module.

[0042] The detail enhancement module is configured to perform detail enhancement processing on the image region in the region mark library having the detail mark by histogram equalization to obtain an enhanced region if the detail mark library has the detail mark.

[0043] The noise reduction module is configured to perform noise reduction processing on the image region in the region mark library having the noise mark by Gaussian filtering to obtain an enhanced region if the noise mark library has the noise mark.

[0044] Optionally, the image fusion module comprises a restored image module and a brightness enhancement module.

[0045] The restored image module is configured to convert the fusion image from the RGB color space to the Lab color space and calculate the correction parameters of the a channel and the b channel at the pixel point (i, j) by a color correction formula and by With The color at the pixel point (i, j) in the a channel and the b channel is corrected to obtain a restored image;

[0046] The luminance enhancement module is configured to adjust the luminance mean value of any pixel point in the restored image and convert the restored image from the Lab color space to the RGB color space to obtain a final enhanced image;

[0047] The color correction formula is:

[0048]

[0049]

[0050] wherein M*N is the image size, D a (i,j) and D b (i,j) are the pixels at (i, j) in the a channel and the b channel, respectively.

[0051] The beneficial effects of the present application are:

[0052] The present application provides an image feature enhancement method for monitoring equipment, which comprises the following steps: obtaining an original image, preprocessing the original image to obtain a luminance mean value, and dividing the original image into multiple image regions; extracting the image features of each image region, adding image markers to the image region according to the image features; storing each image region in the corresponding marker library according to the image markers; enhancing the image regions in the detail marker library and the noise marker library; splicing the enhanced image regions in the coordinate marker library, the detail marker library and the noise marker library according to the coordinates to obtain a fusion image, and adjusting the luminance of the pixel points in the fusion image according to the luminance mean value. This processing method can effectively perform fine processing on different parts of the image, improve the overall quality of the image, and maintain the accuracy of the color and the authenticity of the details. BRIEF DESCRIPTION OF DRAWINGS

[0053] The present application will be further described below with reference to the accompanying drawings.

[0054] Figure 1 A flowchart of an image feature enhancement method for monitoring equipment is provided for the embodiments of the present application.

[0055] Figure 2 Another flowchart of an image feature enhancement method for monitoring equipment is provided for the embodiments of the present application.

[0056] Figure 3 A structural diagram of an image feature enhancement device for monitoring equipment is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of the present application.

[0058] The embodiment of the present application provides an image feature enhancement method for monitoring equipment. Figure 1 , Figure 1 The embodiment of the present application provides a flow chart of an image feature enhancement method for monitoring equipment. The method comprises the following steps:

[0059] S101, an original image is acquired, and the original image is preprocessed to obtain a brightness mean value, and the original image is averagely divided to obtain a plurality of image regions.

[0060] S102, image features of each image region are extracted, and image markers are added to the image region according to the image features.

[0061] S103, each image region is stored in a corresponding marker library according to the image marker.

[0062] S104, the image regions in the detail marker library and the noise marker library are respectively enhanced.

[0063] S105, the enhanced image regions in the coordinate marker library, the detail marker library and the noise marker library are spliced according to coordinates to obtain a fusion image, and the brightness of the pixel points in the fusion image is adjusted according to the brightness mean value.

[0064] The image features include the definition of image details and the number of noise points; the image markers include coordinates, detail markers and noise markers; if the definition of the image region exceeds a preset detail value, the detail marker is added; and if the noise points of the image region exceed a threshold value, the noise marker is added.

[0065] The marker libraries include a coordinate marker library, a detail marker library and a noise marker library; the image regions in the coordinate marker library only contain coordinates, the image regions in the detail marker library contain detail markers, and the image regions in the noise marker library contain noise markers.

[0066] Based on the image feature enhancement method for monitoring equipment provided by the embodiment of the present application, the image is divided into a plurality of regions, and the details and noise points of each region are independently optimized, and finally the overall brightness and color are spliced and adjusted. This processing method can more effectively perform fine processing on different parts of the image, improve the overall quality of the image, and at the same time maintain the accuracy of the color and the authenticity of the details.

[0067] In one implementation, dividing the original image into multiple regions according to the size of the image and optimizing the regions separately improves the efficiency and accuracy of image processing. When the original image is divided, coordinates are added to the image regions, which contain the position information of the image regions in the original image.

[0068] In one implementation, adding labels to the image regions, where the labels of the image regions at least contain one coordinate, or contain two labels including a coordinate and one of a noise label and a detail label, or contain three labels including a coordinate, a noise label and a detail label, improves the accuracy and efficiency of image classification and recognition tasks.

[0069] In one implementation, the images are added to the corresponding coordinate label library, detail label library and noise label library according to the coordinates, detail labels and noise labels of the image regions. By optimizing the image regions in the label library in the corresponding label set, the optimization efficiency can be improved. For example, in the divided image regions, the image regions with only coordinates are added to the coordinate label library, the images with coordinates and noise labels are added to the noise label library, the images with coordinates and detail labels are added to the detail label library, and the images with coordinates, noise labels and detail labels are added to the noise label library and the detail label library, respectively. The coordinate label cannot be modified, and the image region with only coordinates does not need to be enhanced.

[0070] In one implementation, the image regions with three labels in the label library are superimposed and optimized to retain the advantages of the enhanced detail label library and noise label library, and the complete image is obtained by splicing the coordinates of the image itself, which retains the advantages of the enhanced image and facilitates subsequent color correction and image brightness adjustment.

[0071] In one embodiment, step S101 comprises:

[0072] The first data is obtained by converting the original image from the RGB color space to the XYZ color space, the second data is obtained by converting the first data from the XYZ color space to the Lab color space, and the third data is obtained by binarizing the second data.

[0073] The average brightness of the image in the third data is calculated as the brightness mean value.

[0074] In one implementation, converting the original image from the RGB color space to the XYZ color space and then to the Lab color space facilitates more accurate representation and processing of the brightness channel of the image.

[0075] In one implementation, converting the image into the Lab color space can effectively retain a wider color gamut.

[0076] In one embodiment, referring to Figure 2 On the basis of Figure 1 Step S102 further includes:

[0077] S1021, convert the image region from spatial domain to frequency domain, calculate the mean value of the high frequency part in the frequency domain to obtain the definition, and if the definition exceeds the preset detail value, add a detail mark to the image region;

[0078] S1022, obtain the pixel points in the image region that deviate from the value of the surrounding pixel points by more than a threshold value as noise points, and if the number of noise points exceeds a preset number, add a noise point mark to the image region.

[0079] In one implementation, the definition is calculated by a frequency domain analysis method, and a detail mark is added to the image region that exceeds the preset value, which is beneficial to improve the accuracy and visualization effect of image detail information.

[0080] In one implementation, the noise point features are extracted, and a noise point mark is added to the image region that exceeds the threshold value, which is beneficial to improve the efficiency and accuracy of image denoising processing.

[0081] In one embodiment, step S104 further includes:

[0082] If the detail mark library has a detail mark, the image region in the region mark library with the detail mark is subjected to detail enhancement processing by histogram equalization to obtain an enhanced region.

[0083] If the noise point mark library has a noise point mark, the image region in the region mark library with the noise point mark is subjected to noise reduction processing by Gaussian filtering to obtain an enhanced region.

[0084] In one implementation, the use of histogram equalization can enhance the contrast and details of the image, making the image clearer and more full, and improving the visual effect.

[0085] In one implementation, the use of Gaussian filtering can effectively reduce the noise in the image, preserve the image details, and reduce image blur, improving the image quality.

[0086] In one embodiment, step S105 can include the following steps:

[0087] Convert the fusion image from the RGB color space to the Lab color space and calculate the correction parameters of the a channel and the b channel at the pixel point (i, j) by the color correction formula And By And Correct the color of the pixel point (i, j) in the a channel and the b channel to obtain a restored image;

[0088] The final enhanced image is obtained by adjusting any pixel point in the restored image through the brightness mean value and converting the restored image from the Lab color space to the RGB color space.

[0089] Color correction formula:

[0090]

[0091] wherein M*N is the image size, D a (i,j) and D b (i,j) are the pixels at (i,j) in the a channel and the b channel respectively.

[0092] In an implementation, the color accuracy and consistency of the image can be improved by performing color correction on any pixel point in the spliced image through the formula and .

[0093] In an implementation, the brightness of the restored image is corrected by the brightness mean value, and the brightness exceeding the preset threshold value and the brightness lower than the preset threshold value are corrected, so that the brightness of the image is more balanced.

[0094] Based on the same inventive concept, the embodiments of the present application also provide an image feature enhancement device for a monitoring device. Referring to Figure 3 , Figure 3 a framework diagram of an image feature enhancement device for a monitoring device provided by the embodiments of the present application, comprising:

[0095] a preprocessing module configured to acquire an original image, perform preprocessing on the original image to obtain a brightness mean value, and divide the original image into multiple image regions by averaging;

[0096] a label adding module configured to extract an image feature of each image region, and add an image label to the image region according to the image feature; the image feature includes the definition of image details and the number of noise points; the image label includes coordinates, a detail label and a noise point label; if the definition of the image region exceeds a preset detail value, the detail label is added; if the noise points of the image region exceed a threshold value, the noise point label is added;

[0097] a label storage module configured to store each image region in a corresponding label library according to the image label; the label library includes a coordinate label library, a detail label library and a noise point label library; the image regions in the coordinate label library only include coordinates, the image regions in the detail label library include the detail label, and the image regions in the noise point label library include the noise point label;

[0098] an image enhancement module configured to perform enhanced images on the image regions in the detail label library and the noise point label library respectively;

[0099] The image fusion module is used for splicing the enhanced image regions in the coordinate mark library, the detail mark library and the noise mark library according to the coordinates to obtain a fusion image, and adjusting the brightness of the pixel points in the fusion image according to the brightness mean value.

[0100] Based on the image feature enhancement device for monitoring equipment provided by the embodiment of the application, the image is divided into multiple regions, and the details and noise points of each region are independently optimized, and finally the overall brightness and color are spliced and adjusted. This processing method can more effectively perform fine processing on different parts of the image, improve the overall quality of the image, and maintain the accuracy of the color and the authenticity of the details.

[0101] In one embodiment, the preprocessing module includes an image conversion module and a brightness calculation module.

[0102] The image conversion module is used for converting the original image from the RGB color space to the XYZ color space to obtain first data, converting the first data from the XYZ color space to the Lab color space to obtain second data, and performing binaryzation processing on the second data to obtain third data.

[0103] The brightness calculation module is used for calculating the average brightness of the image in the third data as the brightness mean value.

[0104] In one embodiment, the mark adding module includes a detail mark module and a noise point mark module.

[0105] The detail mark module is used for converting the image region from the spatial domain to the frequency domain, calculating the mean value of the high-frequency part in the frequency domain to obtain the definition, and adding a detail mark to the image region if the definition exceeds a preset detail value.

[0106] The noise point mark module is used for obtaining the pixel points in the image region that deviate from the numerical values of the surrounding pixel points by more than a threshold value as noise points, and adding a noise mark to the image region if the number of the noise points exceeds a preset number.

[0107] In one embodiment, the image enhancement module includes a detail enhancement module and a noise reduction module.

[0108] The detail enhancement module is used for performing detail enhancement processing on the image region in the region mark library with the detail mark by histogram equalization to obtain an enhanced region if the detail mark library has the detail mark.

[0109] The noise reduction module is used for performing noise reduction processing on the image region in the region mark library with the noise mark by Gaussian filtering to obtain an enhanced region if the noise mark library has the noise mark.

[0110] In one embodiment, the image fusion module includes a restored image module and a brightness enhancement module.

[0111] The reducing image module is configured to convert the fused image from an RGB color space to a Lab color space and calculate correction parameters of an a channel and a b channel at a pixel point (i, j) by a color correction formula The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is

[0112] The luminance enhancement module is configured to adjust any pixel point in the reducing image by a luminance mean value and inversely convert the reducing image from the Lab color space to the RGB color space to obtain a final enhanced image. The color correction formula is

[0113] The color correction formula is The color correction formula is

[0114] The color correction formula is The color correction formula is The color correction formula is

[0115] The color correction formula is a The color correction formula is b The color correction formula is The color correction formula is

[0116] The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is The color correction formula is

Claims

1. An image feature enhancement method for monitoring equipment, characterized in that, The method includes: The original image is acquired, and the original image is preprocessed to obtain the average brightness value. The original image is then divided into multiple image regions on an average basis. Extract the image features of each image region, and add image labels to the image region based on the image features; Image features include the sharpness of image details and the amount of noise; image markers include coordinates, detail markers, and noise markers. If the sharpness of an image region exceeds a preset detail value, a detail marker is added; if the noise of an image region exceeds a threshold, a noise marker is added. Each image region is stored in a corresponding tag library based on the image tags; the tag library includes a coordinate tag library, a detail tag library, and a noise tag library; the image regions in the coordinate tag library contain only coordinates, the image regions in the detail tag library contain detail tags, and the image regions in the noise tag library contain noise tags; Enhance the image regions in the detail marker library and the noise marker library respectively; The enhanced image regions from the coordinate marker library, detail marker library, and noise marker library are combined according to their coordinates to obtain a fused image. The brightness of the pixels in the fused image is then adjusted based on the average brightness value.

2. The image feature enhancement method for monitoring equipment according to claim 1, characterized in that, Obtain the original image and preprocess it to obtain the average brightness value, including: The first data is obtained by converting the original image from RGB color space to XYZ color space, the second data is obtained by converting the first data from XYZ color space to Lab color space, and the third data is obtained by binarizing the second data. The average brightness of the image in the third dataset is calculated as the brightness mean.

3. The image feature enhancement method for monitoring equipment according to claim 1, characterized in that, Adding image tags to the image region based on image features includes: The image region is transformed from the spatial domain to the frequency domain, and the mean value of the high-frequency part in the frequency domain is calculated to obtain the sharpness. If the sharpness exceeds the preset detail value, a detail marker is added to the image region. Pixels in the image region whose values ​​deviate from those of surrounding pixels by more than a threshold are identified as noise. If the number of noise pixels exceeds a preset number, noise markers are added to the image region.

4. The image feature enhancement method for monitoring equipment according to claim 1, characterized in that, Image enhancement is performed on image regions from the detail marker library and the noise marker library, respectively, including: If the detail marker library contains detail markers, then the image regions in the region marker library containing detail markers are subjected to detail enhancement processing through histogram equalization to obtain enhanced regions; If the noise marker library contains noise markers, then Gaussian filtering is used to perform noise reduction processing on the image regions in the noise marker library to obtain enhanced regions.

5. The image feature enhancement method for monitoring equipment according to claim 1, characterized in that, The method for adjusting the brightness of pixels in a fused image based on the average brightness value includes: The fused image is converted from the RGB color space to the Lab color space, and the correction parameters of the a and b channels at pixel (i, j) are calculated using the color correction formula. and pass and The image is restored by correcting the color at pixel (i, j) in channels a and b. The final enhanced image is obtained by adjusting the average brightness of any pixel in the restored image and inversely converting the restored image from the Lab color space to the RGB color space. The color correction formula is as follows: Where M*N is the image size, D a (i,j) and D b (i,j) represent the pixels at (i,j) in channels a and b, respectively.

6. An image feature enhancement device for monitoring equipment, characterized in that, The device includes a preprocessing module, a marker addition module, a marker storage module, an image enhancement module, and an image fusion module. The preprocessing module is used to acquire the original image, preprocess the original image to obtain the average brightness value, and divide the original image into multiple image regions on an average basis. The labeling module is used to extract image features of each image region and add image labels to the image region based on the image features. Image features include the sharpness of image details and the amount of noise; image markers include coordinates, detail markers, and noise markers. If the sharpness of an image region exceeds a preset detail value, a detail marker is added; if the noise of an image region exceeds a threshold, a noise marker is added. The tag storage module is used to store each image region into a corresponding tag library according to the image tags; the tag library includes a coordinate tag library, a detail tag library, and a noise tag library; the image regions in the coordinate tag library only contain coordinates, the image regions in the detail tag library contain detail tags, and the image regions in the noise tag library contain noise tags. The image enhancement module is used to enhance the image regions in the detail marker library and the noise marker library respectively. The image fusion module is used to combine the enhanced image regions from the coordinate marker library, detail marker library, and noise marker library according to coordinates to obtain a fused image, and adjust the brightness of the pixels in the fused image according to the average brightness value.

7. An image feature enhancement device for monitoring equipment according to claim 6, characterized in that, The preprocessing module includes an image conversion module and a brightness calculation module. The image conversion module is used to obtain first data by converting the original image from RGB color space to XYZ color space, converting the first data from XYZ color space to Lab color space to obtain second data, and performing binarization processing on the second data to obtain third data. The brightness calculation module is used to calculate the average brightness of the image in the third data as the brightness mean.

8. An image feature enhancement device for monitoring equipment according to claim 6, characterized in that, The marker addition module includes a detail marker module and a noise marker module: The detail marking module is used to convert the image region from the spatial domain to the frequency domain, calculate the mean value of the high-frequency part in the frequency domain to obtain the sharpness, and add detail marking to the image region if the sharpness exceeds the preset detail value. The noise marking module is used to identify pixels in the image region whose numerical deviation from the surrounding pixels exceeds a threshold as noise. If the number of noise pixels exceeds a preset number, noise marking is added to the image region.

9. An image feature enhancement device for monitoring equipment according to claim 6, characterized in that, The image enhancement module includes a detail enhancement module and a noise reduction module. The detail enhancement module is used to perform detail enhancement processing on the image regions in the region marker library with detail markers through histogram equalization to obtain enhanced regions if the detail marker library has detail markers. The noise reduction module is used to perform noise reduction processing on the image region in the noise-marked region marker library by Gaussian filtering to obtain an enhanced region if the noise marker library has noise markers.

10. An image feature enhancement device for monitoring equipment according to claim 6, characterized in that, The image fusion module includes an image restoration module and a brightness enhancement module. The image restoration module is used to convert the fused image from the RGB color space to the Lab color space and calculate the correction parameters of the a channel and b channel at pixel (i, j) using a color correction formula. and pass and The image is restored by correcting the color at pixel (i, j) in channels a and b. The brightness enhancement module is used to adjust the average brightness of any pixel in the restored image and inversely convert the restored image from Lab color space to RGB color space to obtain the final enhanced image; The color correction formula is as follows: Where M*N is the image size, D a (i,j) and D b (i,j) represent the pixels at (i,j) in channels a and b, respectively.

Images