Low-light image enhancement method and device, electronic equipment and computer readable medium

Abstract

The embodiment of the invention discloses a low-light image enhancement method and device, electronic equipment and a computer readable medium. One specific embodiment of the method comprises the following steps: collecting an initial low-light image; performing image quality detection processing on the initial low-light image; in response to determining that the image quality detection information does not meet the preset detection condition, sending exposure adjustment information corresponding to the image quality detection information to the low-light camera device so as to adjust the light sensitivity and the exposure time of the low-light camera device, and collecting the low-light image again through the adjusted low-light camera device; acquiring a normal light reference image; inputting the normal light reference image into a reference image feature extraction collaborative network of a pre-trained low-light image enhancement model; and inputting the low-light image and the reference image feature information into the image enhancement association network to obtain an enhanced low-light image corresponding to the low-light image. According to the embodiment, the image quality and the image definition of the enhanced low-light image are improved.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to a low-light image enhancement method, device, electronic device, and computer-readable medium. Background Art

[0002] Low-light image enhancement is a technology for enhancing low-light images captured by a low-light camera (e.g., a low-light camera). Currently, when enhancing low-light images captured by a low-light camera, a method generally adopted is to directly enhance the low-light images captured by the low-light camera.

[0003] However, when the above method is used to enhance the low-light image captured by the low-light camera device, the following technical problems often occur:

[0004] Directly enhancing the low-light images captured by the low-light camera device without optimizing the acquisition process based on the low-light images captured in the low-light environment, the original images captured by the unoptimized acquisition process may contain a lot of noise, and during the enhancement process, the noise will also be amplified at the same time, and the noise of the enhanced low-light image may be more obvious, causing the image to look more grainy and rougher. At the same time, directly enhancing the low-light images captured by the low-light camera device can improve the brightness and contrast of the low-light images to a certain extent, but due to the poor quality of the original images, the quality of the enhanced images may still be limited, resulting in the enhanced low-light images still being relatively blurry.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the invention

[0006] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0007] Some embodiments of the present disclosure provide a low-light image enhancement method, an apparatus, an electronic device, and a computer-readable medium to solve one or more of the technical problems mentioned in the above background technology section.

[0008] In a first aspect, some embodiments of the present disclosure provide a low-light image enhancement method, the method comprising: acquiring an initial low-light image through a low-light camera device; performing image quality detection processing on the initial low-light image to obtain image quality detection information; in response to determining that the image quality detection information satisfies a preset detection condition, determining the initial low-light image as a low-light image; in response to determining that the image quality detection information does not satisfy the preset detection condition, sending exposure adjustment information corresponding to the image quality detection information to the low-light camera device to adjust the sensitivity and exposure time of the low-light camera device, and acquiring a low-light image again through the adjusted low-light camera device; acquiring a normal-light reference image; inputting the normal-light reference image into a reference image feature extraction collaborative network of a pre-trained low-light image enhancement model to obtain reference image feature information corresponding to the normal-light reference image, wherein the low-light image enhancement model includes the reference image feature extraction collaborative network and an image enhancement association network; inputting the low-light image and the reference image feature information into the image enhancement association network to obtain an enhanced low-light image corresponding to the low-light image.

[0009] In a second aspect, some embodiments of the present disclosure provide a low-light image enhancement device, the device comprising: a first acquisition unit, configured to acquire an initial low-light image through a low-light camera device; an image quality detection unit, configured to perform image quality detection processing on the initial low-light image to obtain image quality detection information; a determination unit, configured to determine the initial low-light image as a low-light image in response to determining that the image quality detection information meets a preset detection condition; and a second acquisition unit, configured to send exposure adjustment information corresponding to the image quality detection information to the low-light camera device in response to determining that the image quality detection information does not meet the preset detection condition, so as to adjust the exposure of the low-light camera. The sensitivity and exposure time of the device are adjusted, and the low-light image is captured again by the adjusted low-light camera device; an acquisition unit is configured to acquire a normal-light reference image; a first input unit is configured to input the normal-light reference image into a reference image feature extraction collaborative network of a pre-trained low-light image enhancement model to obtain reference image feature information corresponding to the normal-light reference image, wherein the low-light image enhancement model includes the reference image feature extraction collaborative network and an image enhancement association network; a second input unit is configured to input the low-light image and the reference image feature information into the image enhancement association network to obtain an enhanced low-light image corresponding to the low-light image.

[0010] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the above-mentioned first aspect.

[0011] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner of the above-mentioned first aspect is implemented.

[0012] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the low-light image enhancement method of some embodiments of the present disclosure, the image quality and image clarity of the enhanced low-light image are improved. Specifically, the reason for the poor image quality and image clarity of the enhanced low-light image is that the low-light image captured by the low-light camera device is directly enhanced, and the acquisition process is not optimized according to the low-light image captured in the low-light environment. The original image captured by the unoptimized acquisition process may contain a lot of noise. In the enhancement process, the noise will also be amplified at the same time. The noise of the enhanced low-light image may be more obvious, resulting in a stronger grainy image and a rougher image quality. At the same time, although the low-light image captured by the low-light camera device is directly enhanced, the brightness and contrast of the low-light image can be improved to a certain extent, but due to the poor quality of the original image, the quality of the enhanced image may still be limited, resulting in the enhanced low-light image is still relatively blurred. Based on this, the low-light image enhancement method of some embodiments of the present disclosure firstly captures an initial low-light image through a low-light camera device. Thus, an initial low-light image for determining whether to re-capture the low-light image can be obtained. Then, the image quality detection process is performed on the initial low-light image to obtain image quality detection information. Thus, the image quality detection can be performed on the initial low-light image to generate image quality detection information. The image quality detection information can characterize the quality of the image captured by the low-light camera device without optimizing the acquisition process in a low-light environment. Afterwards, in response to determining that the image quality detection information meets the preset detection condition, the initial low-light image is determined as a low-light image. Thus, on the premise that the image quality detection information of the initial low-light image meets the preset detection condition, the initial low-light image with better image quality can be determined as a low-light image. Then, in response to determining that the image quality detection information does not meet the preset detection condition, the exposure adjustment information corresponding to the image quality detection information is sent to the low-light camera device to adjust the sensitivity and exposure time of the low-light camera device, and the low-light image is captured again by the adjusted low-light camera device. Thus, when the image quality of the initial low-light image is poor (that is, the image quality detection information does not meet the preset detection condition), the sensitivity and exposure time of the low-light camera device can be adjusted, and the parameters of the low-light camera device can be optimized to optimize the acquisition process. Adjusting the sensitivity and exposure time after quality inspection can make the low-light camera device better adapt to the low-light environment, collect low-light images with relatively good quality, and reduce the introduction of noise. Then, a normal light reference image is obtained. Thus, a normal light reference image for generating reference image feature information can be obtained. Next, the normal light reference image is input into the reference image feature extraction collaborative network of a pre-trained low-light image enhancement model to obtain reference image feature information corresponding to the normal light reference image, wherein the low-light image enhancement model includes the reference image feature extraction collaborative network and the image enhancement association network.Therefore, the reference image feature information of the normal light reference image can be extracted through the reference image feature extraction collaborative network. Afterwards, the above-mentioned low-light image and the above-mentioned reference image feature information are input into the above-mentioned image enhancement association network to obtain an enhanced low-light image corresponding to the above-mentioned low-light image. Therefore, the low-light image can be enhanced on the basis of the reference image feature information through the image enhancement association network to obtain an enhanced low-light image. Also, before the low-light image captured by the low-light camera device is enhanced, the quality of the initial low-light image captured by the low-light camera device is detected, the sensitivity and exposure time of the low-light camera device are adjusted, and the parameters of the low-light camera device are optimized to optimize the acquisition process. Adjusting the sensitivity and exposure time after quality detection can enable the low-light camera device to better adapt to the low-light environment, collect low-light images with relatively good quality, reduce the introduction of noise in the low-light image itself, and improve the image quality and image clarity of the enhanced low-light image. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0014] Figure 1 is a flow chart of some embodiments of the low-light image enhancement method according to the present disclosure;

[0015] Figure 2 is a schematic structural diagram of some embodiments of the low-light image enhancement device according to the present disclosure;

[0016] Figure 3 is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure;

[0017] Figure 4 is a schematic diagram of a model structure according to some embodiments of the low-light image enhancement model disclosed herein;

[0018] Figure 5 is another model structure schematic diagram of some embodiments of the low-light image enhancement model according to the present disclosure;

[0019] Figure 6 is a schematic diagram of a network structure of some embodiments of an image enhancement association network according to the present disclosure;

[0020] Figure 7 Schematic diagram of the network structure of some embodiments of the reference image feature extraction collaborative network disclosed in the present invention. DETAILED DESCRIPTION

[0021] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0022] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0023] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0024] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0025] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0026] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0027] Figure 1 A process 100 of some embodiments of the low-light image enhancement method according to the present disclosure is shown. The low-light image enhancement method comprises the following steps:

[0028] Step 101: Capture an initial low-light image using a low-light camera device.

[0029] In some embodiments, the execution subject (e.g., a computing device) of the low-light image enhancement method may capture an initial low-light image through a low-light camera device. The low-light camera device may be an image capture device (e.g., a low-light camera) that captures images in a low-light environment (e.g., at night or in a dimly lit environment). The initial low-light image may be an image captured in a low-light environment when the sensitivity and exposure time of the low-light camera device are not adjusted.

[0030] Step 102: Perform image quality detection processing on the initial low-light image to obtain image quality detection information.

[0031] In some embodiments, the execution entity may perform image quality detection processing on the initial low-light image to obtain image quality detection information.

[0032] In the process of adopting technical solutions to solve the problems mentioned in the background technology, the following problems are often accompanied:

[0033] When performing quality detection on low-light images, the correlation between the color features of RGB low-light images is relatively complex. The numerical changes in the red (R), green (G), and blue (B) channels will simultaneously affect the type, intensity, and brightness of the color. Directly performing quality detection on RGB low-light images through an image quality evaluation model increases the complexity of the model. The model requires more computing resources to perform quality detection on RGB low-light images, which increases the waste of computing resources during the quality detection process.

[0034] Faced with the above technical problems, we decided to adopt the following solutions:

[0035] In some optional implementations of some embodiments, the execution subject may perform image quality detection processing on the initial low-light image through the following steps to obtain image quality detection information:

[0036] In the first step, the area of ​​the initial low-light image is determined as the low-light image division area.

[0037] In the second step, in response to determining that the low-light image division area is divisible by the preset division area, the initial low-light image is divided into initial low-light image blocks with an area of ​​the preset division area and no overlapping images.

[0038] In the third step, in response to determining that the low-light image division area cannot be divided by the preset division area, the initial low-light image is subjected to sliding window division processing based on the preset sliding window length, the preset sliding window width, the preset horizontal sliding step length, and the preset vertical sliding step length to obtain each initial low-light image block. In practice, the execution subject may execute a pre-packaged division method to divide the initial low-light image based on the sliding window length, the sliding window width, the horizontal sliding step length, and the vertical sliding step length to obtain each initial low-light image block. There are overlapping areas in each initial low-light image block segmented by the sliding window.

[0039] In the fourth step, for each of the obtained initial low-light image blocks, perform the following steps:

[0040] In the first sub-step, for each initial low-light image block pixel in the initial low-light image block, the following steps are performed:

[0041] Sub-step 1: Determine the pixel value information of the initial low-light image block pixel as the pixel value information to be converted. The pixel value information may be Rgb data of the initial low-light image block pixel. The pixel value information may include a red pixel value (R), a green pixel value (G), and a blue pixel value (B). For example, the pixel value information may be "(R=255, G=0, B=0)".

[0042] Sub-step two, normalize the above-mentioned information to be converted to obtain normalized pixel value information. In practice, the above-mentioned execution entity may determine the ratio of the red pixel value included in the above-mentioned information to be converted to the preset value as the updated red pixel value, determine the ratio of the green pixel value included in the above-mentioned information to be converted to the preset value as the updated green pixel value, and determine the ratio of the blue pixel value to the preset value as the updated blue pixel value. Afterwards, the above-mentioned execution entity may determine the above-mentioned updated red pixel value, the above-mentioned updated green pixel value, and the above-mentioned updated blue pixel value as the normalized pixel value information. The above-mentioned preset value may be 255. As an example, the above-mentioned information to be converted may be "(R=255, G=0, B=0)", and the above-mentioned normalized pixel value information may be "(R=1, G=0, B=0)".

[0043] Sub-step three: based on the normalized pixel value information, generate pixel hue information corresponding to the pixel of the initial low-light image block. In practice, the execution subject may generate pixel hue information based on the normalized pixel value information by color space conversion technology. As an example, the execution subject may input the updated red pixel value R, the updated green pixel value G, and the updated blue pixel value B into the following formula:

[0044]

[0045] The value range of θ is [0, π]. When B is less than or equal to G, the pixel hue information may be the above θ, and when B is greater than G, the pixel hue information may be the difference between 2π and θ.

[0046] Sub-step 4: Based on the normalized pixel value information, generate pixel saturation information corresponding to the pixel of the initial low-light image block. In practice, the execution subject may generate pixel saturation information based on the normalized pixel value information by color space conversion technology. As an example, the execution subject may input the updated red pixel value R, the updated green pixel value G, and the updated blue pixel value B into the following formula:

[0047]

[0048] When I is equal to 0, the pixel saturation information is 0; when I is not equal to 0, the pixel saturation information can be S, where S can be

[0049] Sub-step 5: Based on the normalized pixel value information, generate the pixel brightness information corresponding to the pixel of the initial low-light image block. In practice, the execution subject may generate the pixel brightness information based on the normalized pixel value information by color space conversion technology. As an example, the execution subject may determine the average of the updated red pixel value R, the updated green pixel value G, and the updated blue pixel value B as the pixel brightness information.

[0050] In a second sub-step, the generated hue information of each pixel point corresponding to each pixel point of the initial low-light image block in the initial low-light image block is determined as the hue information of the initial low-light image block.

[0051] In a third sub-step, the generated saturation of each pixel point corresponding to each initial low-light image block pixel point in the above-mentioned initial low-light image block is determined as the initial low-light image block saturation information.

[0052] In a fourth sub-step, the generated brightness information of each pixel point corresponding to each pixel point of the initial low-light image block in the initial low-light image block is determined as the brightness information of the initial low-light image block.

[0053] In a fifth sub-step, the hue information of the initial low-light image block, the saturation information of the initial low-light image block, and the brightness information of the initial low-light image block corresponding to the initial low-light image block are input into a pre-trained image block quality detection model to obtain image block quality detection score information. The image block quality detection model may be a neural network model (such as a VGGNet (full name Visual Geometry Group Network) model or a Deep Residual Networks (ResNet) model) that takes the hue information of the initial low-light image block, the saturation information of the initial low-light image block, and the brightness information of the initial low-light image block as input and takes the image block quality detection score information as output. The image block quality detection score information may be a score for the image quality. Optionally, the image block quality detection model may be a model for characterizing the correspondence between the hue information of the initial low-light image block, the saturation information of the initial low-light image block, the brightness information of the initial low-light image block, and the image block quality detection score information.

[0054] The fifth step is to generate image quality detection information based on the obtained quality detection score information of each image block. In practice, the execution subject may determine the mean of each score represented by the quality detection score information of each image block as the image quality detection information.

[0055] The above technical solution combines the steps and related contents as an inventive point of the embodiment of the present disclosure, and solves the technical problem of "waste of computing resources". The factors that lead to waste of computing resources are often as follows: when performing quality detection on low-light images, the correlation between the color features of RGB low-light images is relatively complex, and the numerical changes of the red (R), green (G), and blue (B) channels will simultaneously affect the type, intensity, and brightness of the color. The quality detection of RGB low-light images is directly performed on the image quality evaluation model, which increases the complexity of the model. The model requires more computing resources to perform quality detection on RGB low-light images, which increases the waste of computing resources in the quality detection process. At the same time, the human visual system's perception of color is mainly based on hue, saturation, and brightness. However, the organization of the RGB color space does not directly match human visual perception. When performing image quality detection, the analysis results based on the RGB space may be inconsistent with the subjective perception of the human eye. The human eye has an intuitive perception of hue and saturation. For images with good model detection quality but poor perception by the human eye, the image itself may have some color combination or distribution problems, resulting in low accuracy of image quality detection. If the above factors are solved, the effect of reducing the waste of computing resources can be achieved. In order to achieve this effect, first, the area of ​​the initial low-light image is determined as the low-light image division area. Thus, the low-light image division area used to divide the low-light image can be obtained. Then, in response to determining that the low-light image division area is divisible by the preset division area, the initial low-light image is divided into initial low-light image blocks having an area of ​​the preset division area and no overlapping images. Thus, the initial low-light image can be divided into initial low-light image blocks. Afterwards, in response to determining that the low-light image division area is not divisible by the preset division area, the initial low-light image is subjected to sliding window division processing based on the preset sliding window length, the preset sliding window width, the preset horizontal sliding step length, and the preset vertical sliding step length to obtain initial low-light image blocks. Thus, when the low-light image division area is not divisible by the preset division area, the initial low-light image can be divided into initial low-light image blocks. Next, for each of the obtained initial low-light image blocks, the following steps are performed: Step 1, for each initial low-light image block pixel in the above-mentioned initial low-light image block, the following steps are performed: First sub-step, the pixel value information of the above-mentioned initial low-light image block pixel is determined as the pixel value information to be converted. In this way, the pixel value information to be converted for generating normalized pixel value information can be obtained. Second sub-step, normalizing the above-mentioned information to be converted to obtain normalized pixel value information. Third sub-step, based on the above-mentioned normalized pixel value information, generating pixel hue information corresponding to the above-mentioned initial low-light image block pixel. Fourth sub-step, based on the above-mentioned normalized pixel value information, generating pixel saturation information corresponding to the above-mentioned initial low-light image block pixel.The fifth sub-step is to generate the pixel brightness information corresponding to the pixel of the initial low-light image block based on the normalized pixel value information. The second sub-step is to determine the hue information of each pixel corresponding to each initial low-light image block in the initial low-light image block as the hue information of the initial low-light image block. Thus, the initial low-light image block hue information of the initial low-light image block can be obtained. The third sub-step is to determine the saturation of each pixel corresponding to each initial low-light image block in the initial low-light image block as the saturation information of the initial low-light image block. Thus, the initial low-light image block saturation information representing the saturation of the initial low-light image block can be obtained. The fourth sub-step is to determine the brightness information of each pixel corresponding to each initial low-light image block in the initial low-light image block as the brightness information of the initial low-light image block. Thus, the initial low-light image block brightness information representing the brightness of the initial low-light image block can be obtained. In the fifth step, the hue information of the initial low-light image block, the saturation information of the initial low-light image block, and the brightness information of the initial low-light image block corresponding to the initial low-light image block are input into the pre-trained image block quality detection model to obtain the image block quality detection score information. Thus, the hue information (H, i.e., the hue information of the initial low-light image block), the saturation information (S, i.e., the saturation information of the initial low-light image block), and the brightness information (I) of the separated initial low-light image block can be input into the image block quality detection model. This separation makes the data structure clearer, and the model does not need to consider the complex coupling relationship between the RGB channels when processing, thereby reducing the complexity of the data and reducing the consumption of computing resources in the quality detection process. At the same time, the three dimensions of the hue (H), saturation (S), and brightness (I) of the image block are closely related to the human eye's perception of color. When judging the image quality, the human eye mainly depends on the type of color (hue), vividness (saturation), and brightness (brightness). When the HSI information is input into the quality evaluation model, the model's evaluation of image quality is closer to the human eye's perception, which improves the accuracy of image quality detection. Finally, based on the quality detection score information of each image block obtained, image quality detection information is generated. Thus, image quality detection information corresponding to the low-light image can be generated. Also, when performing quality detection on the low-light image, the image data of the RGB low-light image block is converted into initial low-light image block hue information, initial low-light image block saturation information, and initial low-light image block brightness information representing the hue (H), saturation (S), and brightness (I) of the image block. The quality of the image block is detected by using the initial low-light image block hue information, initial low-light image block saturation information, and initial low-light image block brightness information. There is no need to consider the complex coupling relationship between the RGB channels, which reduces the complexity of the data and reduces the consumption of computing resources in the quality detection process.

[0056] Step 103 : in response to determining that the image quality detection information satisfies a preset detection condition, determining the initial low-light image as a low-light image.

[0057] In some embodiments, the execution subject may determine the initial low-light image as a low-light image in response to determining that the image quality detection information satisfies a preset detection condition, wherein the preset detection condition may be greater than or equal to a preset value (score).

[0058] Step 104, in response to determining that the image quality detection information does not meet the preset detection conditions, sending exposure adjustment information corresponding to the image quality detection information to the low-light camera device to adjust the sensitivity and exposure time of the low-light camera device, and capturing the low-light image again through the adjusted low-light camera device.

[0059] In some embodiments, in response to determining that the image quality detection information does not meet the preset detection conditions, the execution subject may send exposure adjustment information corresponding to the image quality detection information to the low-light camera device to adjust the sensitivity and exposure time of the low-light camera device, and collect the low-light image again through the adjusted low-light camera device. The exposure adjustment information may be instruction information for controlling the sensitivity (ISO) and exposure time of the low-light camera device to change the exposure degree of the image. For example, the image quality detection information may be "image quality is 3", and the exposure adjustment information corresponding to "image quality is 3" may be "{"device_id":"001", / / Number of the low-light camera device; "command":"exposure_adjustment", / / Command type is exposure adjustment; "parameters":{"iso":800, / / Increase the sensitivity to 800; "exposure_time":1 / 15 / / Extend the exposure time to 1 / 15 seconds;}}".

[0060] Step 105: Acquire a normal light reference image.

[0061] In some embodiments, the execution subject may obtain a normal light reference image. In practice, the execution subject may obtain the normal light reference image through a wired connection or a wireless connection. The normal light reference image may be an image taken in a normal light environment.

[0062] It should be noted that the above-mentioned wireless connection methods may include but are not limited to 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0063] Step 106 , input the normal light reference image into a reference image feature extraction collaborative network of a pre-trained low light image enhancement model to obtain reference image feature information corresponding to the normal light reference image.

[0064] In some embodiments, the execution subject may input the normal light reference image into a reference image feature extraction collaborative network of a pre-trained low light image enhancement model to obtain reference image feature information corresponding to the normal light reference image. Figure 4 , Figure 5 The low-light image enhancement model includes the reference image feature extraction collaborative network and the image enhancement association network. Figure 7 It is a schematic diagram of the network structure of some embodiments of the reference image feature extraction collaborative network according to the present disclosure. The above-mentioned reference image feature extraction collaborative network includes the above-mentioned reference image feature extraction module, a pooling layer, and a fully connected layer. The above-mentioned reference image feature extraction module includes various convolution modules, and each of the above-mentioned convolution modules includes a convolution layer and an activation function. The above-mentioned low-light image enhancement model can be a neural network model that takes a normal light reference image and a low-light image as input and takes an enhanced low-light image as output. In some optional implementation methods of some embodiments, the above-mentioned execution subject can input the above-mentioned normal light reference image into the reference image feature extraction collaborative network of a pre-trained low-light image enhancement model through the following steps to obtain the reference image feature information corresponding to the above-mentioned normal light reference image:

[0065] The first step is to input the above-mentioned normal light reference image into the reference image feature extraction module of the above-mentioned reference image feature extraction collaborative network to obtain the initial feature information of the reference image. Among them, the above-mentioned reference image feature extraction module includes various convolution modules, and the above-mentioned reference image feature extraction collaborative network includes the above-mentioned reference image feature extraction module, a pooling layer, and a fully connected layer. The above-mentioned reference image feature extraction collaborative network can be a neural network for extracting features from the normal light reference image. Each of the above-mentioned convolution modules includes a convolution layer and an activation function. The output information of the above-mentioned convolution layer points can be input into the above-mentioned activation function layer. The above-mentioned initial feature information of the reference image can be a feature vector or feature map obtained by extracting features from the normal light reference image.

[0066] The initial feature information of the reference image is input into the pooling layer to obtain pooled feature information of the reference image, wherein the reference pooled feature information may be a feature vector or a feature map obtained by performing pooling processing on the initial feature information of the reference image.

[0067] The reference image pooling feature information is input into the fully connected layer to obtain reference image feature information corresponding to the normal light reference image. The reference image feature information may be a higher-level abstract feature vector or a higher-level feature map obtained by processing the reference image pooling feature information through the fully connected layer.

[0068] Step 107 , inputting the feature information of the low-light image and the reference image into an image enhancement association network to obtain an enhanced low-light image corresponding to the low-light image.

[0069] In some embodiments, the execution entity may input the low-light image and the reference image feature information into the image enhancement association network to obtain an enhanced low-light image corresponding to the low-light image.

[0070] In some optional implementations of some embodiments, the execution subject may input the low-light image and the reference image feature information into the image enhancement association network through the following steps to obtain an enhanced low-light image corresponding to the low-light image:

[0071] In the first step, the low-light image is input into the feature extraction module sequence of the image enhancement association network to obtain the feature information of each low-light image. The network structure diagram of the image enhancement association network is shown in FIG. Figure 6 As shown. The above-mentioned image enhancement association network includes the above-mentioned feature extraction module sequence and the feature fusion module sequence, each feature extraction module in the above-mentioned feature extraction module sequence includes a convolution layer and a self-attention block, each feature fusion module in the above-mentioned feature fusion module sequence corresponds to at least one feature extraction module in the above-mentioned feature extraction module sequence, and each feature extraction module in the above-mentioned feature extraction module sequence corresponds to at least one feature fusion module in the feature fusion module sequence. The above-mentioned self-attention block can be a Transformer block. The first feature extraction module in the above-mentioned feature extraction module sequence is configured to perform feature extraction on the low-light image and extract a feature vector representing the context information of the image. Each feature fusion module in the above-mentioned feature fusion module sequence includes a convolution layer and a Transformer block. The above-mentioned feature fusion module can be a decoder. The above-mentioned feature extraction module can be an encoder.

[0072] In the second step, for each feature fusion module in the feature fusion module sequence, at least one low-light image extraction feature information output by at least one feature extraction module corresponding to the feature fusion module in each low-light image extraction feature information, the output information of the previous feature fusion module of the feature fusion module in the feature fusion module sequence, and the reference image feature information are input into the feature fusion module to obtain fused feature information, and the fused feature information is input into the next feature fusion module of the feature fusion module in the feature fusion module sequence. The fused feature information can be a comprehensive representation of feature information from multiple sources (here, at least one low-light image extraction feature information, the output information of the previous feature fusion module, and the reference image feature information). The other fused feature information can be a feature map that combines the features of the low-light image, the accumulated information in the feature fusion process, and the feature information of the normal light image.

[0073] In the third step, the image corresponding to the fused feature information output by the last feature fusion module in the feature fusion module sequence is determined as the enhanced low-light image corresponding to the low-light image. The fused feature information output by the last feature fusion module in the feature fusion module sequence can represent the enhanced low-light image after mapping the multi-level feature information of the normal light reference image to the low-light image. In practice, the execution entity can input the fused feature information output by the last feature fusion module into a preset decoder or preset generator network to map the fused feature information back to the image space to obtain an image corresponding to the fused feature information.

[0074] In some optional implementations of some embodiments, the execution subject may input the low-light image into the feature extraction module sequence of the image enhancement association network to obtain feature information extracted from each low-light image through the following steps:

[0075] In the first step, based on the above feature extraction module sequence, the following steps are performed:

[0076] Sub-step 1: determining each feature extraction module in the feature extraction module sequence except the first feature extraction module as each target feature extraction module.

[0077] Sub-step 2: determining a target feature extraction module located next to the first feature extraction module in the feature extraction module sequence.

[0078] Sub-step three, for the first feature extraction module in the feature extraction module sequence, the low-light image is input into the first feature extraction module to obtain the initial low-light image extraction feature information corresponding to the low-light image. In practice, the convolution layer included in the feature extraction module can perform feature extraction on the low-light image to obtain a feature map representing the low-light image. Then, the self-attention block included in the feature extraction module can receive the feature map representing the low-light image, and for each position (pixel or pixel group) in the low-light image feature map, its feature value will be adjusted according to the calculated attention weight (for example, if the attention weight corresponding to a certain pixel position is higher, then its feature value in the output feature map will be relatively enhanced, and vice versa, if the weight is lower, the feature value will be relatively suppressed.), and the feature map re-weighted by the attention weight is obtained as the initial low-light image extraction feature information.

[0079] Sub-step four: inputting the feature information extracted from the initial low-light image into at least one feature fusion module corresponding to the first feature extraction module.

[0080] Sub-step five, inputting the above-mentioned initial low-light image extraction feature information into the above-mentioned target feature extraction module. Each target feature extraction module in the above-mentioned feature extraction module sequence is configured to receive the output information of the previous feature extraction module of the target feature extraction module in the above-mentioned feature extraction module sequence to output the initial low-light image extraction feature information. The previous feature extraction module of the first target feature extraction module in the feature extraction module sequence may be the first feature extraction module in the feature extraction module sequence. For each target feature extraction module in the feature extraction module sequence except the first target feature extraction module, the previous feature extraction module of the target feature extraction module may be the previous target feature extraction module of the above-mentioned target feature extraction module in the feature extraction module sequence.

[0081] Sub-step six: determining each initial low-light image extraction feature information output by the first feature extraction module in the feature extraction module sequence and each target feature extraction module as each low-light image extraction feature information.

[0082] Optionally, the low-light image enhancement model can be trained by the following steps:

[0083] The first step is to obtain a sample set. The samples in the sample set include sample low-light reference image data and sample target normal-light images corresponding to the sample low-light reference image data. The sample low-light reference image data includes a sample low-light image and a sample normal-light reference image.

[0084] In the second step, the following training steps are performed based on the sample set:

[0085] In a first sub-step, the sample low-light reference image data of at least one sample in the sample set is input into an initial neural network to obtain a sample predicted normal-light image corresponding to each sample in the at least one sample.

[0086] The second sub-step is to compare the sample predicted normal light image corresponding to each sample in the at least one sample with the corresponding sample target normal light image. In practice, the execution subject may compare the sample predicted normal light image corresponding to each sample in the at least one sample with the corresponding sample target normal light image by using a cross entropy loss function to determine the difference between the sample predicted normal light image corresponding to each sample in the at least one sample and the corresponding sample target normal light image.

[0087] The third sub-step is to determine whether the initial neural network has reached a preset optimization target based on the comparison result. The above optimization target may be that the cross entropy loss function value is less than or equal to a preset value.

[0088] In a fourth sub-step, in response to determining that the initial neural network achieves the above-mentioned optimization goal, the initial neural network is used as a trained low-light image enhancement model.

[0089] The fifth sub-step is, in response to determining that the initial neural network does not achieve the above optimization goal, adjusting the network parameters of the initial neural network, using unused samples to form a sample set, using the adjusted initial neural network as the initial neural network, and performing the above training step again. As an example, the network parameters of the above initial neural network can be adjusted using a back propagation algorithm (BP algorithm) and a gradient descent method (such as a mini-batch gradient descent algorithm).

[0090] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the low-light image enhancement method of some embodiments of the present disclosure, the image quality and image clarity of the enhanced low-light image are improved. Specifically, the reason for the poor image quality and image clarity of the enhanced low-light image is that the low-light image captured by the low-light camera device is directly enhanced, and the acquisition process is not optimized according to the low-light image captured in the low-light environment. The original image captured by the unoptimized acquisition process may contain a lot of noise. In the enhancement process, the noise will also be amplified at the same time. The noise of the enhanced low-light image may be more obvious, resulting in a stronger grainy image and a rougher image quality. At the same time, although the low-light image captured by the low-light camera device is directly enhanced, the brightness and contrast of the low-light image can be improved to a certain extent, but due to the poor quality of the original image, the quality of the enhanced image may still be limited, resulting in the enhanced low-light image is still relatively blurred. Based on this, the low-light image enhancement method of some embodiments of the present disclosure firstly captures an initial low-light image through a low-light camera device. Thus, an initial low-light image for determining whether to re-capture the low-light image can be obtained. Then, the image quality detection process is performed on the initial low-light image to obtain image quality detection information. Thus, the image quality detection can be performed on the initial low-light image to generate image quality detection information. The image quality detection information can characterize the quality of the image captured by the low-light camera device without optimizing the acquisition process in a low-light environment. Afterwards, in response to determining that the image quality detection information meets the preset detection condition, the initial low-light image is determined as a low-light image. Thus, on the premise that the image quality detection information of the initial low-light image meets the preset detection condition, the initial low-light image with better image quality can be determined as a low-light image. Then, in response to determining that the image quality detection information does not meet the preset detection condition, the exposure adjustment information corresponding to the image quality detection information is sent to the low-light camera device to adjust the sensitivity and exposure time of the low-light camera device, and the low-light image is captured again by the adjusted low-light camera device. Thus, when the image quality of the initial low-light image is poor (that is, the image quality detection information does not meet the preset detection condition), the sensitivity and exposure time of the low-light camera device can be adjusted, and the parameters of the low-light camera device can be optimized to optimize the acquisition process. Adjusting the sensitivity and exposure time after quality inspection can make the low-light camera device better adapt to the low-light environment, collect low-light images with relatively good quality, and reduce the introduction of noise. Then, obtain a normal light reference image. In this way, a normal light reference image for generating reference image feature information can be obtained. Next, the normal light reference image is input into the reference image feature extraction collaborative network of the pre-trained low-light image enhancement model to obtain the reference image feature information corresponding to the normal light reference image. Among them, the low-light image enhancement model includes the reference image feature extraction collaborative network and the image enhancement association network.Therefore, the reference image feature information of the normal light reference image can be extracted through the reference image feature extraction collaborative network. Afterwards, the above-mentioned low-light image and the above-mentioned reference image feature information are input into the above-mentioned image enhancement association network to obtain an enhanced low-light image corresponding to the above-mentioned low-light image. Therefore, the low-light image can be enhanced on the basis of the reference image feature information through the image enhancement association network to obtain an enhanced low-light image. Also, before the low-light image captured by the low-light camera device is enhanced, the quality of the initial low-light image captured by the low-light camera device is detected, the sensitivity and exposure time of the low-light camera device are adjusted, and the parameters of the low-light camera device are optimized to optimize the acquisition process. Adjusting the sensitivity and exposure time after quality detection can enable the low-light camera device to better adapt to the low-light environment, collect low-light images with relatively good quality, reduce the introduction of noise in the low-light image itself, and improve the image quality and image clarity of the enhanced low-light image.

[0091] Further references Figure 2 As an implementation of the methods shown in the figures, the present disclosure provides some embodiments of a low-light image enhancement device. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0092] like Figure 2As shown, the low-light image enhancement device 200 of some embodiments includes: a first acquisition unit 201, an image quality detection unit 202, a determination unit 203, a second acquisition unit 204, an acquisition unit 205, a first input unit 206, and a second input unit 207. The first acquisition unit 201 is configured to acquire an initial low-light image through a low-light camera device; the image quality detection unit 202 is configured to perform image quality detection processing on the initial low-light image to obtain image quality detection information; the determination unit 203 is configured to determine the initial low-light image as a low-light image in response to determining that the image quality detection information meets a preset detection condition; the second acquisition unit 204 is configured to send exposure adjustment information corresponding to the image quality detection information to the low-light camera device in response to determining that the image quality detection information does not meet the preset detection condition, so as to adjust the sensitivity and exposure time of the low-light camera device, and A low-light image is captured again through the adjusted low-light camera device; the acquisition unit 205 is configured to acquire a normal-light reference image; the first input unit 206 is configured to input the above-mentioned normal-light reference image into a reference image feature extraction collaborative network of a pre-trained low-light image enhancement model to obtain reference image feature information corresponding to the above-mentioned normal-light reference image, wherein the above-mentioned low-light image enhancement model includes the above-mentioned reference image feature extraction collaborative network and the image enhancement association network; the second input unit 207 is configured to input the above-mentioned low-light image and the above-mentioned reference image feature information into the above-mentioned image enhancement association network to obtain an enhanced low-light image corresponding to the above-mentioned low-light image.

[0093] It is understood that the units described in the device 200 are similar to those described in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the device 200 and the units contained therein, and will not be described in detail here.

[0094] Reference below Figure 3 , which shows a structural schematic diagram of an electronic device 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0095] like Figure 3As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0096] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0097] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the functions defined in the method of some embodiments of the present disclosure are executed.

[0098] It should be noted that the computer-readable medium recorded in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device or device. In some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0099] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0100] The computer-readable medium may be included in the electronic device; or may exist separately without being incorporated into the electronic device. The computer-readable medium carries one or more programs, and when one or more programs are executed by the electronic device, the electronic device: collects an initial low-light image through a low-light camera device; performs image quality detection processing on the initial low-light image to obtain image quality detection information; in response to determining that the image quality detection information satisfies a preset detection condition, determines the initial low-light image as a low-light image; in response to determining that the image quality detection information does not satisfy the preset detection condition, sends exposure adjustment information corresponding to the image quality detection information to the low-light camera device to adjust the sensitivity and exposure time of the low-light camera device, and collects a low-light image again through the adjusted low-light camera device; obtains a normal-light reference image; inputs the normal-light reference image into a reference image feature extraction collaborative network of a pre-trained low-light image enhancement model to obtain reference image feature information corresponding to the normal-light reference image, wherein the low-light image enhancement model includes the reference image feature extraction collaborative network and an image enhancement association network; inputs the low-light image and the reference image feature information into the image enhancement association network to obtain an enhanced low-light image corresponding to the low-light image.

[0101] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0102] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0103] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be provided in a processor, for example, may be described as: a processor comprising a first acquisition unit, an image quality detection unit, a determination unit, a second acquisition unit, an acquisition unit, a first input unit, and a second input unit. The names of these units do not, in some cases, constitute limitations on the units themselves, for example, the first acquisition unit may also be described as a "unit for acquiring an initial low-light image through a low-light camera device".

[0104] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0105] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of technical features, but should also cover other technical solutions formed by any combination of technical features or their equivalent features without departing from the inventive concept. For example, a technical solution formed by replacing the features with (but not limited to) the technical features with similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A low-light image enhancement method, comprising: Collecting an initial low-light image by using a low-light camera device; Performing image quality detection processing on the initial low-light image to obtain image quality detection information; In response to determining that the image quality detection information satisfies a preset detection condition, determining the initial low-light image as a low-light image; In response to determining that the image quality detection information does not meet a preset detection condition, sending exposure adjustment information corresponding to the image quality detection information to the low-light camera device to adjust the sensitivity and exposure time of the low-light camera device, and capturing a low-light image again through the adjusted low-light camera device; Acquire a normal light reference image; Inputting the normal light reference image into a reference image feature extraction collaborative network of a pre-trained low light image enhancement model to obtain reference image feature information corresponding to the normal light reference image, wherein the low light image enhancement model includes the reference image feature extraction collaborative network and an image enhancement association network; The low-light image and the reference image feature information are input into the image enhancement association network to obtain an enhanced low-light image corresponding to the low-light image.

2. The method according to claim 1, wherein: The step of inputting the normal light reference image into a reference image feature extraction collaborative network of a pre-trained low light image enhancement model to obtain reference image feature information corresponding to the normal light reference image includes: Inputting the normal light reference image into the reference image feature extraction module of the reference image feature extraction collaborative network to obtain initial feature information of the reference image, wherein the reference image feature extraction module includes various convolution modules, and the reference image feature extraction collaborative network includes the reference image feature extraction module, a pooling layer, and a fully connected layer; Inputting the initial feature information of the reference image into the pooling layer to obtain pooled feature information of the reference image; The reference image pooling feature information is input into the fully connected layer to obtain reference image feature information corresponding to the normal light reference image.

3. The method according to claim 1, wherein: The step of inputting the low-light image and the reference image feature information into the image enhancement association network to obtain an enhanced low-light image corresponding to the low-light image includes: Inputting the low-light image into the feature extraction module sequence of the image enhancement association network to obtain feature information extracted from each low-light image, wherein the image enhancement association network includes the feature extraction module sequence and a feature fusion module sequence, each feature extraction module in the feature extraction module sequence includes a convolution layer and a self-attention block, each feature fusion module in the feature fusion module sequence corresponds to at least one feature extraction module in the feature extraction module sequence, and each feature extraction module in the feature extraction module sequence corresponds to at least one feature fusion module in the feature fusion module sequence; For each feature fusion module in the feature fusion module sequence, at least one low-light image extraction feature information output by at least one feature extraction module corresponding to the feature fusion module in each low-light image extraction feature information, output information of a previous feature fusion module of the feature fusion module in the feature fusion module sequence, and the reference image feature information are input into the feature fusion module to obtain fused feature information, and the fused feature information is input into the next feature fusion module of the feature fusion module in the feature fusion module sequence; An image corresponding to the fused feature information output by the last feature fusion module in the feature fusion module sequence is determined as an enhanced low-light image corresponding to the low-light image.

4. The method according to claim 3, wherein: The step of inputting the low-light image into the feature extraction module sequence of the image enhancement association network to obtain feature information extracted from each low-light image includes: Based on the feature extraction module sequence, the following steps are performed: Determine each feature extraction module in the feature extraction module sequence except the first feature extraction module as each target feature extraction module; Determine a target feature extraction module located at a next position of the first feature extraction module in the feature extraction module sequence; For a first feature extraction module in the feature extraction module sequence, inputting a low-light image into the first feature extraction module to obtain initial low-light image extraction feature information corresponding to the low-light image; Inputting the initial low-light image extracted feature information into at least one feature fusion module corresponding to the first feature extraction module; Inputting the initial low-light image extraction feature information into the target feature extraction module, wherein each target feature extraction module in the feature extraction module sequence is configured to receive output information of a previous feature extraction module of the target feature extraction module in the feature extraction module sequence to output the initial low-light image extraction feature information; The individual initial low-light image extraction feature information output by the first feature extraction module in the feature extraction module sequence and the individual target feature extraction modules are determined as the individual low-light image extraction feature information.

5. The method according to claim 1, wherein: The low-light image enhancement model is trained by the following steps: Acquire a sample set, wherein the samples in the sample set include sample low-light reference image data and a sample target normal-light image corresponding to the sample low-light reference image data, wherein the sample low-light reference image data includes the sample low-light image and the sample normal-light reference image; Perform the following training steps based on the sample set: Inputting sample low-light reference image data of at least one sample in the sample set into an initial neural network to obtain a sample predicted normal-light image corresponding to each sample in the at least one sample; comparing a sample predicted normal light image corresponding to each sample of the at least one sample with a corresponding sample target normal light image; Determine whether the initial neural network reaches the preset optimization goal according to the comparison result; In response to determining that the initial neural network achieves the optimization goal, using the initial neural network as a trained low-light image enhancement model; In response to determining that the initial neural network does not achieve the optimization goal, the network parameters of the initial neural network are adjusted, and unused samples are used to form a sample set, and the adjusted initial neural network is used as the initial neural network to perform the training step again.

6. A low-light image enhancement device, comprising: A first acquisition unit is configured to acquire an initial low-light image through a low-light camera device; an image quality detection unit, configured to perform image quality detection processing on the initial low-light image to obtain image quality detection information; a determining unit configured to determine the initial low-light image as a low-light image in response to determining that the image quality detection information satisfies a preset detection condition; a second acquisition unit, configured to, in response to determining that the image quality detection information does not meet a preset detection condition, send exposure adjustment information corresponding to the image quality detection information to the low-light camera device to adjust the sensitivity and exposure time of the low-light camera device, and to acquire a low-light image again through the adjusted low-light camera device; an acquisition unit configured to acquire a normal light reference image; a first input unit configured to input the normal light reference image into a reference image feature extraction collaborative network of a pre-trained low light image enhancement model to obtain reference image feature information corresponding to the normal light reference image, wherein the low light image enhancement model includes the reference image feature extraction collaborative network and an image enhancement association network; The second input unit is configured to input the low-light image and the reference image feature information into the image enhancement association network to obtain an enhanced low-light image corresponding to the low-light image.

7. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.

8. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images