Image quality evaluation method, electronic device and storage medium

Abstract

The present application relates to the field of image processing and discloses an image quality assessment method, electronic device, and storage medium. The method includes: performing foreground identification on a speckle pattern to obtain the foreground region of the speckle pattern; counting the proportion of void pixels in the region corresponding to the foreground region in a depth map corresponding to the speckle pattern; and evaluating the quality of the depth map based on the void ratio. This solution only counts the void ratio corresponding to foreground objects when evaluating the depth map quality based on the void ratio, making the void ratio statistics more reliable and robust, thereby more accurately reflecting the depth quality assessment.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of image processing, and in particular to an image quality evaluation method, electronic device, and storage medium. Background Art

[0002] Structured light cameras, which generate depth data by emitting active infrared light, are widely used in 3D facial recognition and in a wide range of scenarios, such as payment, door locks, and rail transit. In these scenarios, people are the main subject of the image, while the background is mostly empty, even beyond the camera's operating distance. A key metric for evaluating the quality of depth maps and determining whether the camera structure has been deformed is the depth map's void ratio. The conventional void ratio calculation method uses the denominator as the total number of pixels in the depth map, and the numerator as the number of pixels with a depth value of 0.

[0003] However, in actual use, when the speckle pattern used to generate the depth map has a large background, the depth recovery algorithm cannot recover a high-quality depth map, resulting in an increase in the number of pixels with a depth value of 0 in the depth map and an increase in the error in the calculated void ratio. This makes it impossible to truly evaluate the quality of the depth map, which will cause misjudgment of the business. Summary of the Invention

[0004] The purpose of the embodiments of the present application is to provide an image quality evaluation method, an electronic device and a storage medium. When evaluating the quality of a depth map based on the hole rate, only the hole rate corresponding to the foreground object is counted, so that the hole rate statistics are more reliable and robust, and thus more realistically reflect the depth quality evaluation.

[0005] To solve the above technical problems, an embodiment of the present application provides an image quality evaluation method, comprising:

[0006] performing foreground recognition on the speckle pattern to obtain a foreground area of ​​the speckle pattern;

[0007] Counting the proportion of hollow pixels in an area corresponding to the foreground area in the depth map corresponding to the speckle map;

[0008] The quality of the depth map is evaluated according to the proportion.

[0009] An embodiment of the present application also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above-mentioned image quality evaluation method.

[0010] An embodiment of the present application further provides a computer-readable storage medium storing a computer program, which implements the above-mentioned image quality evaluation method when executed by a processor.

[0011] Compared to existing technologies, the embodiments of the present invention perform foreground recognition on a speckle pattern to obtain the foreground region of the speckle pattern; then calculate the percentage of void pixels in the foreground region of the depth map corresponding to the speckle pattern; and then evaluate the depth map quality based on this percentage. This solution distinguishes between the foreground and background areas of the speckle pattern, allowing for greater focus on the quality of the foreground object when evaluating the depth map quality based on the void ratio. By only calculating the void ratio corresponding to the foreground object, this makes the void ratio statistics more reliable and robust, thereby more accurately reflecting the depth quality assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] One or more embodiments are exemplarily described by the figures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments.

[0013] Figure 1 is a flow chart of the image quality evaluation method provided in an embodiment of the present application;

[0014] Figure 2 is a flow chart of the image quality evaluation method provided in an embodiment of the present application;

[0015] Figure 3 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, each embodiment of the present application will be described in detail below with reference to the accompanying drawings. However, it will be understood by those skilled in the art that in each embodiment of the present application, many technical details are proposed to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present application can be implemented. The division of the following embodiments is for convenience of description and should not constitute any limitation on the specific implementation of the present application. The various embodiments can be combined and referenced with each other under the premise of no contradiction.

[0017] One embodiment of the present application relates to an image quality evaluation method for evaluating the quality of a depth map. The execution subject may be a terminal, device or server capable of image processing, such as Figure 1 As shown, the image quality evaluation method specifically includes the following steps.

[0018] Step 101: perform foreground recognition on the speckle pattern to obtain the foreground area of ​​the speckle pattern.

[0019] Specifically, when using a depth camera to capture a target object and obtain a speckle pattern, the target object is usually considered to be the main subject of the speckle pattern, namely the foreground area, while the rest of the background is usually relatively empty, namely the background area. After obtaining the speckle pattern, image recognition and other technologies can be used to perform foreground recognition (i.e., foreground and background segmentation) on the speckle pattern to obtain the foreground area of ​​the speckle pattern.

[0020] The method used for foreground recognition can be based on factors such as the size, shape, and image reflectivity (expressed as the size and distribution of pixel grayscale) of the target object in the foreground area. This embodiment does not limit the specific method of foreground recognition.

[0021] Step 102 : Count the proportion of hole pixels in the area corresponding to the foreground area in the depth map corresponding to the speckle image.

[0022] Specifically, after obtaining the speckle pattern, the depth camera can perform a disparity search based on the speckle pattern and the camera's reference image to obtain a disparity map corresponding to the speckle pattern; then, for the disparity value of each pixel in the disparity map, the corresponding depth value is calculated using the triangulation principle to form a depth map corresponding to the speckle pattern.

[0023] After obtaining the depth map corresponding to the speckle pattern, the area in the depth map corresponding to the foreground area of ​​the speckle pattern is determined, and the total number of pixels and the number of void pixels in this area are counted. Based on the total number of pixels and the number of void pixels in this area, the proportion of void pixels in the area, i.e., the void ratio, can be calculated.

[0024] Among them, a hole pixel point can be defined as a pixel point with a depth value of 0 or a pixel point with an invalid value.

[0025] Step 103: Evaluate the quality of the depth map according to the proportion.

[0026] Specifically, when evaluating depth map quality, a smaller depth map hole ratio indicates higher image quality, while a larger hole ratio indicates lower image quality. However, in this embodiment, the hole ratio in the depth map region corresponding to the foreground area, which is of greater interest, is used to evaluate the image quality of the entire depth map. This ensures that the hole ratio used is more reliable and robust, thereby more realistically reflecting the depth quality evaluation of the depth map.

[0027] For example, a proportion threshold can be set. When the proportion of empty pixels in the area corresponding to the foreground area in the depth map is less than the proportion threshold, it is considered that the quality of the depth map is high, the availability of the depth map is high, and it can truly reflect the depth information of the target object; when the proportion of empty pixels in the area corresponding to the foreground area in the depth map is not less than the proportion threshold, it is considered that the quality of the depth map is low and cannot truly reflect the depth information of the target object.

[0028] Compared to related technologies, this embodiment performs foreground recognition on the speckle pattern to obtain the foreground area of ​​the speckle pattern; then calculates the percentage of void pixels in the foreground area of ​​the depth map corresponding to the speckle pattern; and evaluates the depth map quality based on this percentage. This solution distinguishes between the foreground and background areas of the speckle pattern, allowing for greater focus on the quality of the foreground objects when evaluating the depth map quality based on the void ratio. By only calculating the void ratio corresponding to the foreground objects, this makes the void ratio statistics more reliable and robust, and thus more accurately reflects the depth quality assessment.

[0029] Another embodiment of the present invention relates to an image quality evaluation method, such as Figure 2 As shown in the figure, the image quality evaluation method is to Figure 1 The improvement of the method steps shown is that the process of obtaining the foreground area of ​​the speckle pattern is refined. Figure 2 As shown, the above step 101 may include the following sub-steps.

[0030] Sub-step 1011: performing overexposure detection on the speckle pattern to determine non-overexposed pixels.

[0031] Specifically, a speckle pattern is generated by a speckle projector that emits coherent light from a laser diode in a structured light camera, focusing it through a lens and dispersing it into multiple random light rays through a diffraction optical element, thereby producing a random dot pattern. The infrared camera senses the pattern projected onto the surface of the object. When the speckle pattern is generated, the exposure of each pixel on the generated speckle pattern will be different due to the different distances between the object surface and the structured light camera. Image overexposure usually refers to an area in the image where the grayscale is 255. In this case, the area is determined to be an overexposed area, and the pixels in the overexposed area are determined to be overexposed pixels. Other areas outside the area are determined to be non-overexposed areas, and the pixels in the non-overexposed areas are determined to be non-overexposed pixels.

[0032] In one example, determining non-overexposed pixels may be achieved through the following steps.

[0033] Step 1: For the pixel with a gray value of 255 in the speckle image, a second window is constructed with the pixel as the center.

[0034] Specifically, for a pixel with a grayscale value of 255 in the speckle image, a two-dimensional window can be constructed with the pixel as the center, which is recorded as the second window; the size of the second window can be customized, such as a 5×5 window with 25 pixels in the window.

[0035] Step 2: When the grayscale standard deviation of a pixel in the second window is less than a first threshold, the pixel is determined as an overexposed pixel, and the remaining pixels in the speckle image except the overexposed pixel are determined as non-overexposed pixels.

[0036] Specifically, the grayscale average μ of the pixels in the second window is obtained, μ=(ΣI i ) / N, where I i is the grayscale value of the i-th pixel in the second window, the value range of i is 1, 2, ..., N, and N is the total number of pixels in the second window. After obtaining the grayscale average value of the second window, the grayscale standard deviation σ of the pixels in the second window is calculated according to the standard deviation formula, σ={[Σ(I i -μ)] 1 / 2 A first threshold is set to measure the grayscale standard deviation boundary value of the second window when the overexposed pixel is used as the center point of the second window. When the grayscale standard deviation of the pixels in the second window is less than the first threshold, it indicates that the grayscale value distribution of the pixels in the second window is relatively balanced, and the center pixel in the second window is determined as an overexposed pixel. When the grayscale standard deviation of the pixels in the second window is not less than the first threshold, it indicates that the grayscale value distribution of the pixels in the second window is relatively uneven, and the center pixel in the second window is not determined as an overexposed pixel. After all overexposed pixels in the speckle pattern are determined, the remaining pixels in the speckle pattern except the overexposed pixels are determined as non-overexposed pixels.

[0037] For the convenience of labeling, the overexposed pixels in the speckle image can be marked as 0 and the non-overexposed pixels can be marked as 1, thereby forming a labeled image Mask1.

[0038] Sub-step 1012: For each non-overexposed pixel, construct a first window with the non-overexposed pixel as the center.

[0039] Specifically, for each non-overexposed pixel in the speckle pattern (i.e., the pixel marked as 1 in Mask1), a two-dimensional window can be constructed with each pixel as the center, which is recorded as the first window. The size of the first window can be customized, such as a 9×9 window with 81 pixels in the window.

[0040] Sub-step 1013: performing similarity matching between the pre-built template and the first window, and determining non-overexposed pixels whose matching results are greater than a similarity threshold as pixels in the foreground area;

[0041] The template is a preset image block with reflectivity characteristics of the foreground area.

[0042] In a structured light camera, coherent light emitted by a laser diode is focused by a lens and dispersed into multiple random rays by a diffractive optical element, thereby producing a speckle projector with a random dot pattern. An infrared camera senses the pattern projected onto the surface of an object and generates a speckle image. Because the energy of each emitted ray varies, the energy at the center of the ray is highest when reflected from the target object. Therefore, each speckle block has a brightest pixel. The speckle brightness follows a two-dimensional normal distribution (a two-dimensional Gaussian distribution) centered around the brightest point. Due to the varying reflectivity of surface materials, the speckle pattern exhibits three different patterns: one in which localized areas retain bright features (i.e., high reflectivity), such as the human body; the second in which localized areas exhibit darker features (i.e., low reflectivity), such as the skin of a face; and the third in which the background area exhibits no usable speckle features, where the energy decays quadratically with the distance from the light source.

[0043] According to the different reflectivities of the foreground and background areas to speckle light, an image block that meets the reflectivity characteristics of the foreground area can be pre-constructed as a template. This template is then used to perform sliding matching in the non-overexposed area of ​​the speckle pattern, thereby determining whether the area crossed in the non-overexposed area is the foreground area.

[0044] Specifically, the window size of the pre-built template can be the same as the first window size, such as a 9×9 window with 81 pixels in the window. This template is then matched with each of the first windows. When the similarity is high, that is, the matching result is greater than a preset similarity threshold, the non-overexposed pixel in the center of the first window is determined to be a pixel in the foreground area. After all pixels in the speckle pattern belonging to the foreground area are determined, the foreground area in the speckle pattern can be determined.

[0045] In one example, when similarity matching of a pre-built template with the first window is performed, the method may include:

[0046] The normalized square difference between the template and the first window is calculated using the following formula, and the obtained normalized square difference is used as the matching result of similarity matching:

[0047]

[0048] Where R(x,y) is the normalized square difference, (x,y) is the coordinate of the non-overexposed pixel in the speckle pattern, (x',y') is the coordinate of the pixel in the template, K(x',y') is the grayscale value of the pixel at (x',y'), and I(x+x',y+y') is the grayscale value of the pixel at (x+x',y+y').

[0049] Generally, when performing similarity matching, the coordinate position of the non-overexposed pixel should be aligned with the coordinate position of the center pixel of the template.

[0050] In one example, the process of building a template may include:

[0051] An image block centered at a central pixel and having a grayscale value obeying a two-dimensional Gaussian distribution is constructed as a template; the standard deviation of the two-dimensional Gaussian function corresponding to the two-dimensional Gaussian distribution is set so that the template has the reflectivity characteristics of the foreground area.

[0052] Specifically, based on the previously analyzed refractive index characteristics of speckle light on the target object in the foreground area, when reflected from the target object, the energy at the center of the ray is highest. Each speckle block has a brightest speckle pixel, and the speckle brightness follows a two-dimensional normal distribution (a two-dimensional Gaussian distribution) centered around the brightest point. Therefore, the template constructed in this embodiment, modeled on this characteristic, is also an image block centered around the central pixel, with grayscale values ​​following a two-dimensional Gaussian distribution. This allows for better matching calculations between the template and the first window described above. To ensure that the template retains the reflectivity characteristics of the foreground area, this embodiment achieves this by adjusting the standard deviation of the two-dimensional Gaussian function corresponding to the two-dimensional Gaussian distribution.

[0053] The specific formula of the two-dimensional Gaussian function is given below:

[0054]

[0055] Among them, σ is the standard deviation, and (x, y) corresponds to the coordinates of the pixel points in the template.

[0056] Based on the previously analyzed refractive index characteristics of speckle light projected onto target objects in the foreground area, we know that the refractive index characteristics projected onto different target objects (such as the human body and skin) vary, necessitating further differentiation based on the target object being photographed. Therefore, when setting templates with foreground reflectivity characteristics, multiple templates can be used, each with a different reflectivity characteristic for the foreground area. Differentiating reflectivity characteristics can be achieved by adjusting the standard deviation of the two-dimensional Gaussian function used by the templates.

[0057] In one example, the template may include a first template, the standard deviation of the two-dimensional Gaussian function corresponding to the first template is smaller than the first standard deviation, and the first template has a high reflectivity feature of the foreground area; the similarity threshold corresponding to the first template is recorded as the high reflectivity threshold.

[0058] On this basis, the above-mentioned process of performing similarity matching between the pre-built template and the first window and determining the non-overexposed pixels whose matching results are greater than the similarity threshold as pixels in the foreground area may include:

[0059] A similarity matching is performed between the pre-constructed first template and the first window, and non-overexposed pixel points whose matching results are greater than a high reflectivity threshold are determined as pixel points in the foreground area.

[0060] Specifically, as mentioned above, due to the varying reflectivity of the surface materials of an object, three types of patterns can be observed. One of these patterns is that localized areas of the speckle pattern retain a characteristic of highlighting, i.e., high reflectivity, such as the human body. The standard deviation of the two-dimensional Gaussian function corresponding to a high-reflectivity template is typically small, meaning that the coefficient at the center of the template is large and the coefficients at the periphery are small. Therefore, a template can be constructed such that the standard deviation of the two-dimensional Gaussian function corresponding to the template is less than a first standard deviation (the first standard deviation is relatively small). This allows the template to retain the high reflectivity characteristics of the foreground area. In this embodiment, the template with the high reflectivity characteristics of the foreground area is referred to as the first template. The similarity threshold used for the first template is denoted as the high reflectivity threshold.

[0061] Accordingly, when performing similarity matching, the first template and the first window can be similarly matched, and the non-overexposed pixel point at the center of the first window with a matching result greater than the high reflectivity threshold is determined as a pixel point in the foreground area.

[0062] For the convenience of annotation, the non-overexposed pixel points in Mask1 whose matching results determined by matching with the first template are greater than the high reflectivity threshold can be further marked as 2, thereby forming an annotated image Mask2.

[0063] In another example, the above template may include a second template, the standard deviation of the two-dimensional Gaussian function corresponding to the second template is greater than the second standard deviation, and the second template has a low reflectivity feature of the foreground area; the similarity threshold corresponding to the second template is recorded as the low reflectivity threshold.

[0064] On this basis, the above-mentioned process of performing similarity matching between the pre-built template and the first window and determining the non-overexposed pixels whose matching results are greater than the similarity threshold as pixels in the foreground area may include:

[0065] A similarity matching is performed between the pre-constructed second template and the first window, and non-overexposed pixel points whose matching results are greater than a low reflectivity threshold are determined as pixel points in the foreground area.

[0066] Specifically, as mentioned above, due to the varying reflectivity of the surface materials of an object, three patterns can be observed. The second is that localized areas of the speckle pattern exhibit darker characteristics, resulting in reduced contrast and, in other words, low reflectivity, such as the skin area of ​​a human face. Low-reflectivity templates typically have a larger standard deviation of the two-dimensional Gaussian function, meaning that the central coefficient of the template is similar to the surrounding coefficients. Therefore, a template can be constructed such that the standard deviation of the two-dimensional Gaussian function corresponding to the template is greater than the second standard deviation (the second standard deviation is relatively large and greater than the first standard deviation). This allows the template to exhibit the low reflectivity characteristics of the foreground area. In this embodiment, the template exhibiting the low reflectivity characteristics of the foreground area is referred to as the second template. The similarity threshold used for the second template is denoted as the low reflectivity threshold.

[0067] Accordingly, when performing similarity matching, the second template can be similarly matched with the first window, and the non-overexposed pixel point at the center of the first window with a matching result greater than the low reflectivity threshold is determined as a pixel point in the foreground area.

[0068] For the convenience of annotation, the non-overexposed pixel points in Mask1 whose matching results determined by matching with the second template are greater than the low reflectivity threshold can be further marked as 3, thereby forming an annotated image Mask3.

[0069] Finally, the pixels in the foreground area that are finally determined may include the pixels marked as 2 in the above Mask2 and the pixels marked as 3 in Mask3.

[0070] Compared to related technologies, this embodiment performs overexposure detection on the speckle pattern to identify non-overexposed pixels. For each non-overexposed pixel, a first window is constructed with the non-overexposed pixel as the center. A similarity match is performed between a pre-constructed template and the first window, and non-overexposed pixels whose matching results exceed a similarity threshold are determined as pixels in the foreground area. The template is a pre-set image block with the reflectivity characteristics of the foreground area. This method allows for rapid identification of pixels in the speckle pattern that belong to the foreground area, facilitating the calculation of the hole rate corresponding to foreground objects when evaluating depth map quality based on the hole rate. This makes hole rate statistics more reliable and robust, and ultimately reflects the ultimate goal of depth quality assessment.

[0071] The present application also relates to an electronic device, such as Figure 3As shown, it includes: at least one processor 201; and a memory 202 that is communicatively connected to the at least one processor 201; wherein the memory 202 stores instructions that can be executed by the at least one processor 201, and the instructions are executed by the at least one processor 201 to enable the at least one processor 201 to execute the image quality evaluation method in the above-mentioned embodiments.

[0072] The memory and processor are connected using a bus, which can include any number of interconnected buses and bridges. The bus connects various circuits of one or more processors and memories. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits. These are all well known in the art and are therefore not described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single component or multiple components, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over a wireless medium via an antenna. Furthermore, the antenna receives data and transmits it to the processor.

[0073] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0074] The embodiments of the present application relate to a computer-readable storage medium storing a computer program, which implements the above-mentioned method embodiments when executed by a processor.

[0075] That is, those skilled in the art will understand that all or part of the steps in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a program, which is stored in a storage medium and includes a number of instructions for causing a device (which may be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc., various media that can store program code.

[0076] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present application, and that in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present application.

Claims

1. A method for evaluating image quality, characterized in that: The method comprises: performing foreground recognition on the speckle pattern to obtain a foreground area of ​​the speckle pattern; Counting the proportion of hollow pixels in an area corresponding to the foreground area in the depth map corresponding to the speckle map; Evaluating the quality of the depth map according to the proportion; The performing foreground recognition on the speckle pattern to obtain the foreground area of ​​the speckle pattern includes: Performing overexposure detection on the speckle pattern to determine non-overexposed pixels; For each non-overexposed pixel, construct a first window with the non-overexposed pixel as the center; Performing similarity matching between the pre-built template and the first window, and determining the non-overexposed pixel points whose matching results are greater than a similarity threshold as pixel points in the foreground area; The template is a preset image block with reflectivity characteristics of the foreground area.

2. The image quality evaluation method according to claim 1, wherein: The performing overexposure detection on the speckle pattern to determine non-overexposed pixels includes: For a pixel point with a grayscale value of 255 in the speckle image, a second window is constructed with the pixel point as the center; When the grayscale standard deviation of a pixel in the second window is less than a first threshold, the pixel is determined as an overexposed pixel, and the remaining pixels in the speckle image except the overexposed pixel are determined as non-overexposed pixels.

3. The image quality evaluation method according to claim 1, wherein: The similarity matching of the pre-built template with the first window includes: The normalized square difference between the template and the first window is calculated using the following formula, and the obtained normalized square difference is used as the matching result of the similarity matching: Wherein, R(x,y) is the normalized square difference, (x,y) is the coordinate of the non-overexposed pixel in the speckle pattern, (x',y') is the coordinate of the pixel in the template, K(x',y') is the grayscale value of the pixel at (x',y'), and I(x+x',y+y') is the grayscale value of the pixel at (x+x',y+y').

4. The image quality evaluation method according to claim 1, wherein: The process of constructing the template includes: Constructing an image block centered at a central pixel and having grayscale values ​​obeying a two-dimensional Gaussian distribution as the template; The standard deviation of the two-dimensional Gaussian function corresponding to the two-dimensional Gaussian distribution is set so that the template has the reflectivity characteristics of the foreground area.

5. The image quality evaluation method according to claim 4, wherein: The template includes a first template, the standard deviation of the two-dimensional Gaussian function corresponding to the first template is smaller than the first standard deviation, and the first template has a high reflectivity feature of the foreground area; The similarity threshold corresponding to the first template is recorded as a high reflectivity threshold; The similarity matching of the pre-built template with the first window and determining the non-overexposed pixel points whose matching results are greater than a similarity threshold as pixel points in the foreground area includes: The pre-constructed first template is similarly matched with the first window, and the non-overexposed pixel points whose matching results are greater than the high reflectivity threshold are determined as pixel points in the foreground area.

6. The image quality evaluation method according to claim 4 or 5, characterized in that: The template includes a second template, the standard deviation of the two-dimensional Gaussian function corresponding to the second template is greater than the second standard deviation, and the second template has the low reflectivity feature of the foreground area; The similarity threshold corresponding to the second template is recorded as a low reflectivity threshold; The similarity matching of the pre-built template with the first window and determining the non-overexposed pixel points whose matching results are greater than a similarity threshold as pixel points in the foreground area includes: The pre-constructed second template is similarly matched with the first window, and the non-overexposed pixel points whose matching results are greater than the low reflectivity threshold are determined as pixel points in the foreground area.

7. The image quality evaluation method according to claim 1, wherein: The hole pixel point is a pixel point with a depth value of 0 or a pixel point with an invalid value.

8. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the image quality assessment method according to any one of claims 1 to 7.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the image quality evaluation method according to any one of claims 1 to 7 is implemented.

Images