High-precision CMOS image sensor data storage optimization method and device based on adaptive compression algorithm, medium and equipment

Abstract

The invention relates to a high-precision CMOS image sensor data storage optimization method and device based on an adaptive compression algorithm, a medium and equipment, and relates to the technical field of image processing, and the method comprises the steps: determining the personalized perception configuration of a user for an image based on the behavior data of the user under the condition of user permission; acquiring a target image acquired by the high-precision CMOS image sensor; generating a perception weight map associated with the target image according to the personalized perception configuration; segmenting the sensing weight map into a plurality of areas, and calculating the weight of each area; according to the weight of each region, determining the importance level of the region; and compressing a part associated with the region in the target image according to the importance level. In this way, the image storage and transmission efficiency is remarkably improved, and meanwhile the personalized requirement of a user for the image quality is met.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing technologies, and in particular, to a method, device, medium, and equipment for optimizing the data storage of a high-precision CMOS image sensor based on an adaptive compression algorithm. Background Art

[0002] With the wide application of high-precision CMOS (Complementary Metal-Oxide-Semiconductor) image sensors, the resolution and quality of image data have been significantly improved, but the resulting data storage and transmission pressures have also become increasingly large. Traditional image compression methods are difficult to achieve a balance among user requirements, compression efficiency, and image quality. Summary of the Invention

[0003] The purpose of the present disclosure is to provide a method, device, medium, and equipment for optimizing the data storage of a high-precision CMOS image sensor based on an adaptive compression algorithm to solve the above technical problems.

[0004] To achieve the above purpose, in a first aspect, the present disclosure provides a method for optimizing the data storage of a high-precision CMOS image sensor based on an adaptive compression algorithm, including: Determining a personalized perception configuration of the user for an image based on the user's behavior data with the user's permission; Obtaining a target image collected by a high-precision CMOS image sensor; Generating a perception weight map associated with the target image according to the personalized perception configuration; Dividing the perception weight map into multiple regions and calculating the weights of each region; Determining the importance level of the region according to the weights of each region; Compressing the part of the target image associated with the region according to the importance level.

[0005] Optionally, the determining the personalized perception configuration of the user for an image based on the user's behavior data includes: Conducting a brightness sensitivity test on the user to obtain the user's brightness sensitivity to an image; Conducting a color sensitivity test on the user to obtain the user's color sensitivity to an image; Conducting a contrast sensitivity test on the user to obtain the user's contrast sensitivity to an image; Conducting a texture complexity preference test on the user to obtain the user's texture complexity preference for an image; Wherein, the personalized perception configuration includes the brightness sensitivity, color sensitivity, contrast sensitivity, and texture complexity preference.

[0006] Optionally, generating a perception weight map associated with the target image according to the personalized perception configuration includes: Generating a user perception vector according to the personalized perception configuration, where the user perception vector includes a brightness sensitivity dimension, a color sensitivity dimension, a contrast sensitivity dimension, and a texture complexity preference dimension; Generating a feature map of the target image, where each pixel in the feature map includes a brightness feature, a color feature, a contrast feature, and a texture feature; For the feature map, performing weighted superposition of each pixel in the feature map with the user perception vector to obtain the perception weight map.

[0007] Optionally, splitting the perception weight map into multiple regions and calculating the weights of each region includes: Splitting the perception weight map into multiple regions; For each region, calculating the average weight of the pixel points within the region; Taking the average weight as the weight of the region.

[0008] Optionally, determining the importance level of the region according to the weights of each region includes: For each region, when the weight of the region is greater than a first threshold, determining the importance level of the region as the first level; When the weight of the region is greater than a second threshold and less than or equal to the first threshold, determining the importance level of the region as the second level; When the weight of the region is greater than a third threshold and less than or equal to the second threshold, determining the importance level of the region as the third level; Wherein, the third threshold is less than the second threshold, the second threshold is less than the first threshold, and the importance degrees of the first level, the second level, and the third level decrease in sequence.

[0009] Optionally, compressing the part of the target image associated with the region according to the importance level includes: When the importance level of the region is the first level, performing a first degree of compression on the part of the target image associated with the region; When the importance level of the region is the second level, performing a second degree of compression on the part of the target image associated with the region; When the importance level of the region is the third level, performing a third degree of compression on the part of the target image associated with the region; Among them, at the first level, the second level, and the third level, the compression degree of the part of the target image associated with the region gradually increases.

[0010] Optionally, the image is a JPEG image. The first-level compression of the part of the target image associated with the region includes: compressing the part of the target image associated with the region with a quality factor of 10-30. The second-level compression of the part of the target image associated with the region includes: compressing the part of the target image associated with the region with a quality factor of 50-70. The third-level compression of the part of the target image associated with the region includes: compressing the part of the target image associated with the region with a quality factor of 90-100.

[0011] In a second aspect, a high-precision CMOS image sensor data storage optimization device based on an adaptive compression algorithm is provided, including: A first module, configured to determine the personalized perception configuration of the user for the image based on the user's behavior data with the user's permission. A second module, configured to acquire a target image collected by a high-precision CMOS image sensor. A third module, configured to generate a perception weight map associated with the target image according to the personalized perception configuration. A fourth module, configured to divide the perception weight map into multiple regions and calculate the weights of each region. A fifth module, configured to determine the importance level of the region according to the weights of each region. A sixth module, configured to compress the part of the target image associated with the region according to the importance level.

[0012] In a third aspect, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the steps of the method described in any item of the first aspect are implemented.

[0013] In a fourth aspect, a device is provided, including: A memory, on which a computer program is stored. A processor, configured to execute the computer program in the memory to implement the steps of the method described in any item of the first aspect.

[0014] In the above solution, personalized perception weight maps can be generated in combination with user behavior data, so as to reflect the degree of attention of users to different regions of the image. In this way, in combination with the adaptive compression algorithm, the image can be differentially compressed according to the importance level of the region, achieving a balance between high-fidelity storage of key regions and efficient compression of non-key regions. Ultimately, the efficiency of image storage and transmission is significantly improved, while meeting the personalized requirements of users for image quality.

[0015] Other features and advantages of the present disclosure will be described in detail in the following specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. They are used together with the following specific implementation to explain the present disclosure, but do not constitute a limitation to the present disclosure. In the drawings: Figure 1 is a flowchart of a method for optimizing data storage of a high-precision CMOS image sensor based on an adaptive compression algorithm.

[0017] Figure 2 is a device diagram of a method for optimizing data storage of a high-precision CMOS image sensor based on an adaptive compression algorithm.

[0018] Figure 3 is a block diagram of an electronic device. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The following will describe in detail the specific implementation of the present disclosure with reference to the drawings. It should be understood that the specific implementation described herein is only for explaining and understanding the present disclosure, and is not used to limit the present disclosure.

[0020] With the wide application of high-precision CMOS image sensors, the resolution and quality of image data have been significantly improved, but the subsequent data storage and transmission pressures have also become increasingly large. Traditional image compression methods often cannot balance compression efficiency and image quality. Especially when there are personalized differences in the attention points of users to different regions of the image, a unified compression strategy may lead to detail loss in key regions or waste of storage resources. Therefore, there is an urgent need for a solution that can optimize image storage according to the personalized perception characteristics of users, so as to significantly reduce data storage requirements while ensuring the quality of key regions.

[0021] To achieve the above object, the present disclosure provides a method for optimizing data storage of a high-precision CMOS image sensor based on an adaptive compression algorithm. Figure 1 is a flowchart of a method for optimizing data storage of a high-precision CMOS image sensor based on an adaptive compression algorithm. Referring to Figure 1 , the method includes: S11. With the user's permission, determine the user's personalized perception configuration for images based on the user's behavior data; S12. Obtain a target image collected by a high-precision CMOS image sensor; S13. Generate a perception weight map associated with the target image according to the personalized perception configuration; S14. Divide the perception weight map into multiple regions and calculate the weights of each region; S15. Determine the importance level of the region according to the weights of each region; S16. Compress the part of the target image associated with the region according to the importance level.

[0022] The implementation manners of the above steps are exemplarily described below.

[0023] In step S11, with the user's permission, determine the user's personalized perception configuration for images based on the user's behavior data.

[0024] In one implementation manner, the determining the user's personalized perception configuration for images based on the user's behavior data includes: Conduct a brightness sensitivity test on the user to obtain the user's brightness sensitivity to images; Conduct a color sensitivity test on the user to obtain the user's color sensitivity to images; Conduct a contrast sensitivity test on the user to obtain the user's contrast sensitivity to images; Conduct a texture complexity preference test on the user to obtain the user's texture complexity preference for images; Wherein, the personalized perception configuration includes the brightness sensitivity, color sensitivity, contrast sensitivity, and texture complexity preference.

[0025] For example, in the brightness sensitivity test, a series of grayscale images with different brightness levels can be shown to the user, and the brightness of these images gradually changes from low to high (for example, the grayscale value ranges from 10 to 250). The user can mark the brightness range in which details are most easily distinguishable in an image containing different brightness regions. Or the user can select the most comfortable brightness in multiple scene images with different brightness levels. Among them, different brightness levels correspond to different brightness sensitivities.

[0026] If the user can clearly distinguish details in the low-brightness region (such as grayscale value 10 - 50), it indicates that the user is sensitive to low brightness. If the user prefers to select images in the high-brightness region (such as grayscale value 200 - 250), it indicates that the user is more sensitive to high brightness. In this way, the user's brightness sensitivity can be determined, and when compressing the image, the information in the corresponding brightness region can be preferentially retained based on the requirements.

[0027] During the color sensitivity test, a set of color gradient images can be shown to the user for testing hue, saturation, and lightness respectively: Hue test: The user observes a gradient image from red to blue and marks the area where the color change is most easily perceived.

[0028] Saturation test: The user observes a set of images with saturation changing from low to high and selects the saturation range that is easiest for them to distinguish.

[0029] Brightness - color combination test: The user observes a set of color images at different brightness levels and selects the brightness - color combination they prefer.

[0030] In this way, the user's choices in the hue, saturation, and lightness tests can be recorded, and the user's sensitive areas for colors can be analyzed. For example, if the user is more sensitive to changes in the light blue area (such as the part with lower saturation in a blue gradient), it indicates that the user pays more attention to light blue and low - saturation areas. If the user is more sensitive to changes in high - saturation colors (such as red or green), it indicates that the user pays more attention to strong color contrasts. Based on the test results, the user's color sensitivity can be determined.

[0031] During the contrast sensitivity test, a set of images with different contrasts can be shown to the user. The test is divided into two parts: Edge sharpness test: The user observes an image containing areas with different contrasts (such as an edge image from blurred to clear) and marks the edge area that is easiest for them to distinguish.

[0032] Contrast preference test: The user observes a set of natural - scene images with contrasts changing from low to high and selects the contrast range they prefer.

[0033] In this way, the marked areas in the edge sharpness test and the user's choices in the contrast preference test can be recorded. If the user can more easily distinguish high - contrast areas (such as the clear - edge part), it indicates that the user is more sensitive to high contrasts. If the user prefers low - contrast scene images, it indicates that the user pays more attention to images with smooth transitions. Based on the test results, the contrast sensitivity can be determined.

[0034] During the texture complexity preference test, a set of images with different texture complexities can be shown to the user. The test is divided into two parts: Texture discrimination ability test: The user observes an image containing simple textures (such as regular lines) and complex textures (such as interlaced patterns or natural textures) and marks the texture area that is easier for them to distinguish.

[0035] Texture preference test: The user observes a set of images with varying texture complexities from low to high and selects the texture type they prefer more.

[0036] In this way, the marked areas in the texture discrimination ability test and the selections in the texture preference test of the user can be recorded. If the user pays more attention to complex texture areas (such as high-frequency interleaved patterns), it indicates that the user is more interested in complex textures. If the user prefers simple texture areas (such as regular lines), it indicates that the user is more concerned about low-frequency textures. According to the test results, the user's texture complexity preference can be determined.

[0037] In this way, the personalized perception configuration includes the brightness sensitivity, color sensitivity, contrast sensitivity, and texture complexity preference. The image can be personalized compressed according to the personalized perception configuration.

[0038] In step S12, obtain the target image collected by the high-precision CMOS image sensor.

[0039] In step S13, generate a perception weight map associated with the target image according to the personalized perception configuration.

[0040] In a possible implementation manner, the generating a perception weight map associated with the target image according to the personalized perception configuration includes: Generate a user perception vector according to the personalized perception configuration, where the user perception vector includes a brightness sensitivity dimension, a color sensitivity dimension, a contrast sensitivity dimension, and a texture complexity preference dimension.

[0041] Continuing with the above example, the brightness sensitivity, color sensitivity, contrast sensitivity, and texture complexity preference can be normalized to generate a user perception vector. As an example, the user perception vector can be: U = [0.4, 0.3, 0.2, 0.1].

[0042] Among them, the brightness sensitivity weight is 0.4, biased towards low-brightness areas. Color sensitivity: The weight is 0.3, biased towards light blue areas. Contrast sensitivity: The weight is 0.2, biased towards high-contrast areas. Texture complexity preference: The weight is 0.1, biased towards complex texture areas.

[0043] In addition, a feature map of the target image can be generated, and each pixel in the feature map includes a brightness feature, a color feature, a contrast feature, and a texture feature. For the feature map, each pixel in the feature map is weighted and superimposed with the user perception vector to obtain the perception weight map.

[0044] Exemplarily, for the input target image, multiple brightness feature maps, color feature maps, contrast feature maps, and texture feature maps can be extracted.

[0045] In the brightness feature map, the brightness distribution of each pixel in the target image can be reflected. In the color feature map, the color components of each pixel in the target image (such as the L, a, b values in the CIELAB color space) can be described. In the contrast feature map, high-contrast regions can be extracted through edge detection algorithms. In the texture feature map, the complexity of the texture in the image can be extracted through filters (such as Gabor filters).

[0046] In this way, each feature map reflects the information of the image in a specific dimension. Combining the feature maps, each pixel point can have brightness features, color features, contrast features, and texture features.

[0047] In this way, for the feature maps, each pixel in the feature maps can be weighted and superimposed with the user perception vector to obtain the perception weight map. For example, the brightness feature of the pixel can be multiplied by the brightness dimension in the user perception vector, and the color feature of the pixel can be multiplied by the color dimension in the user perception vector. Similarly, the contrast feature and the texture feature can also be processed, and finally the perception weight map is obtained.

[0048] In step S14, the perception weight map is divided into multiple regions, and the weights of each region are calculated.

[0049] Exemplarily, the perception weight map can be divided into blocks of a fixed size (such as 8×8 pixels). Similarly, since the perception weight map corresponds to the target image, the target image can also be divided into blocks of a fixed size (such as 8×8 pixels). The blocks obtained by dividing the perception weight map and the blocks obtained by dividing the target image are in one-to-one correspondence.

[0050] Optionally, the dividing the perception weight map into multiple regions and calculating the weights of each region includes:[[]] Dividing the perception weight map into multiple regions; For each region, calculating the average value of the weights of the pixel points in the region; Taking the average value of the weights as the weight of the region.

[0051] That is to say, the values of the pixel points in the region in the perception weight map can be averaged to obtain the weight of the region.

[0052] In step S15, according to the weights of each region, the importance level of the region is determined.

[0053] In one implementation manner, the determining the importance level of the region according to the weights of each region includes:[[]] For each region, when the weight of the region is greater than a first threshold, determine that the importance level of the region is the first level; When the weight of the region is greater than a second threshold and less than or equal to the first threshold, determine that the importance level of the region is the second level; When the weight of the region is greater than a third threshold and less than or equal to the second threshold, determine that the importance level of the region is the third level; Wherein, the third threshold is less than the second threshold, the second threshold is less than the first threshold, and the importance degrees of the first level, the second level, and the third level decrease in sequence.

[0054] It should be noted that the first threshold, the second threshold, and the third threshold can be set according to requirements. When the weight is greater than the first threshold, it indicates that in the user's perception, the importance degree of the region is the highest (the easiest to perceive), so the importance level of the region is the first level. When the weight is greater than the second threshold and less than or equal to the first threshold, it indicates that in the user's perception, the importance degree of the region is relatively high, but not as high as the first level. Therefore, the importance level of the region is the second level. When the weight is greater than the third threshold and less than or equal to the second threshold, it indicates that in the user's perception, the importance degree of the region is the lowest, so the importance level of the region is the third level.

[0055] In step S16, compress the part of the target image associated with the region according to the importance level.

[0056] In one implementation manner, the compressing the part of the target image associated with the region according to the importance level includes: When the importance level of the region is the first level, perform a first degree of compression on the part of the target image associated with the region; When the importance level of the region is the second level, perform a second degree of compression on the part of the target image associated with the region; When the importance level of the region is the third level, perform a third degree of compression on the part of the target image associated with the region; Wherein, under the first degree, the second degree, and the third degree, the compression degree of the part of the target image associated with the region gradually increases.

[0057] As an example, the image is a JPEG image. When the importance level of the region is the first level, the performing a first degree of compression on the part of the target image associated with the region includes: compressing the part of the target image associated with the region with a quality factor of 10 - 30.

[0058] When the importance level of the said region is the second level, the second-degree compression of the part of the target image associated with the said region includes: compressing the part of the target image associated with the said region with a quality factor of 50 - 70; When the importance level of the said region is the third level, the third-degree compression of the part of the target image associated with the said region includes: compressing the part of the target image associated with the said region with a quality factor of 90 - 100.

[0059] In this way, in JPEG, different regions with different importance levels can be compressed respectively through quantization tables of different levels (for example, discrete cosine transform and quantization are performed block by block for compression). For regions with higher importance levels that are easily perceptible to users, less compression is performed. For regions with lower importance levels that are not easily perceptible to users, more compression is performed.

[0060] In the above solution, personalized perception weight maps can be generated by combining user behavior data, so as to reflect the attention degree of users to different regions of the image. In this way, combined with the adaptive compression algorithm, the image can be differentially compressed according to the importance level of the region, achieving a balance between high-fidelity storage of key regions and efficient compression of non-key regions. Finally, the efficiency of image storage and transmission is significantly improved, while meeting the personalized requirements of users for image quality.

[0061] Of course, in the final compression result, the compressed image data (the compression results of all image blocks), the region importance level (facilitating restoration during decoding), and the weight perception map can be retained, which can be stored as additional data and used to guide image reconstruction during decoding.

[0062] Based on the same inventive concept, a high-precision CMOS image sensor data storage optimization device based on an adaptive compression algorithm is also provided. Figure 2 It is a block diagram of a high-precision CMOS image sensor data storage optimization device based on an adaptive compression algorithm. Refer to Figure 2 , the said device includes: The first module 201 is used to determine the personalized perception configuration of the user for the image based on the behavior data of the user with the permission of the user; The second module 202 is used to acquire the target image collected by the high-precision CMOS image sensor; The third module 203 is used to generate a perception weight map associated with the target image according to the personalized perception configuration; The fourth module 204 is used to divide the perception weight map into multiple regions and calculate the weights of each region; The fifth module 205 is configured to determine the importance level of the region according to the weights of the respective regions; The sixth module 206 is configured to compress the part associated with the region in the target image according to the importance level.

[0063] In the above solution, a personalized perception weight map can be generated in combination with user behavior data, so as to reflect the degree of attention of the user to different regions of the image. In this way, in combination with an adaptive compression algorithm, the image can be differentially compressed according to the importance level of the region, achieving a balance between high-fidelity storage of key regions and efficient compression of non-key regions. Finally, the efficiency of image storage and transmission is significantly improved, while meeting the personalized needs of users for image quality.

[0064] In one implementation, the first module is configured to: Conduct a brightness sensitivity test on the user to obtain the user's brightness sensitivity to the image; Conduct a color sensitivity test on the user to obtain the user's color sensitivity to the image; Conduct a contrast sensitivity test on the user to obtain the user's contrast sensitivity to the image; Conduct a texture complexity preference test on the user to obtain the user's texture complexity preference for the image; Wherein, the personalized perception configuration includes the brightness sensitivity, color sensitivity, contrast sensitivity, and texture complexity preference.

[0065] In one implementation, the third module is configured to: Generate a user perception vector according to the personalized perception configuration, where the user perception vector includes a brightness sensitivity dimension, a color sensitivity dimension, a contrast sensitivity dimension, and a texture complexity preference dimension; Generate a feature map of the target image, where each pixel in the feature map includes a brightness feature, a color feature, a contrast feature, and a texture feature; For the feature map, perform weighted superposition of each pixel in the feature map with the user perception vector to obtain the perception weight map.

[0066] In one implementation, the fourth module is configured to: Segment the perception weight map into multiple regions; For each region, calculate the weighted average value of the pixel points in the region; Use the weighted average value as the weight of the region.

[0067] In one implementation, the fifth module is configured to: For each region, when the weight of the region is greater than the first threshold, determine that the importance level of the region is the first level; When the weight of the region is greater than the second threshold and less than or equal to the first threshold, determine that the importance level of the region is the second level; When the weight of the region is greater than the third threshold and less than or equal to the second threshold, determine that the importance level of the region is the third level; Wherein, the third threshold is less than the second threshold, the second threshold is less than the first threshold, and the importance degrees of the first level, the second level, and the third level decrease in sequence.

[0068] In one embodiment, the sixth module is configured to: When the importance level of the region is the first level, perform a first degree of compression on the part of the target image associated with the region; When the importance level of the region is the second level, perform a second degree of compression on the part of the target image associated with the region; When the importance level of the region is the third level, perform a third degree of compression on the part of the target image associated with the region; Wherein, under the first degree, the second degree, and the third degree, the compression degree of the part of the target image associated with the region gradually increases.

[0069] In one embodiment, the image is a JPEG image, and performing a first degree of compression on the part of the target image associated with the region includes: compressing the part of the target image associated with the region with a quality factor of 10 - 30; Performing a second degree of compression on the part of the target image associated with the region includes: compressing the part of the target image associated with the region with a quality factor of 50 - 70; Performing a third degree of compression on the part of the target image associated with the region includes: compressing the part of the target image associated with the region with a quality factor of 90 - 100.

[0070] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of the method of the present invention are implemented.

[0071] The present invention also provides a device, including: A memory, on which a computer program is stored; A processor for executing the computer program in the memory to implement the steps of the above method of the present invention.

[0072] Regarding the device in the above embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be elaborated here.

[0073] Figure 3 is a block diagram of an electronic device 700 shown according to an exemplary embodiment. As Figure 3 shown, the electronic device 700 may include: a processor 701, a memory 702. The electronic device 700 may further include one or more of a multimedia component 703, an input / output (I / O) interface 704, and a communication component 705.

[0074] Among them, the processor 701 is used to control the overall operation of the electronic device 700 to complete all or part of the steps in the above method. The memory 702 is used to store various types of data to support the operation of the electronic device 700. These data may include, for example, instructions for any application or method operating on the electronic device 700, as well as application-related data, such as contact data, received and sent messages, pictures, audio, video, and so on. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc. The multimedia component 703 may include a screen and an audio component. Among them, the screen may be a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone, and the microphone is used to receive external audio signals. The received audio signals may be further stored in the memory 702 or sent through the communication component 705. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 704 provides an interface between the processor 701 and other interface modules. The above other interface modules may be a keyboard, a mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 705 is used for wired or wireless communication between the electronic device 700 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G, or 4G, or a combination of one or more of them. Therefore, the corresponding communication component 705 may include: a Wi-Fi module, a Bluetooth module, and an NFC module.

[0075] In an exemplary embodiment, the electronic device 700 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the above method.

[0076] In another exemplary embodiment, a computer-readable storage medium including program instructions is further provided. When the program instructions are executed by a processor, the steps of the above method are implemented. For example, the computer-readable storage medium may be the memory 702 including the program instructions, and the above program instructions may be executed by the processor 701 of the electronic device 700 to complete the above method.

[0077] In another exemplary embodiment, a computer program product is further provided. The computer program product includes a computer program that can be executed by a processor. When the computer program is executed by the processor, the steps of the above method are implemented.

[0078] The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings. However, the present disclosure is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present disclosure, various simple modifications can be made to the technical solutions of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0079] In addition, it should be noted that, in the above specific embodiments, the various specific technical features described can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the present disclosure will not separately describe various possible combination manners.

[0080] Furthermore, any combination can be made between various different embodiments of the present disclosure as long as it does not violate the idea of the present disclosure, and it should also be regarded as the content disclosed by the present disclosure.

Claims

1. A high-precision CMOS image sensor data storage optimization method based on an adaptive compression algorithm, characterized in that: include: With the permission of the user, determining the user's personalized perception configuration of the image based on the user's behavior data; Acquire the target image captured by the high-precision CMOS image sensor; generating a perceptual weight map associated with the target image according to the personalized perceptual configuration; Dividing the perception weight map into a plurality of regions and calculating the weight of each region; According to the weight of each area, determine the importance level of the area; A portion of the target image associated with the region is compressed according to the importance level.

2. The method according to claim 1, characterized in that The determining, based on the user's behavior data, the user's personalized perception configuration of the image includes: Conducting a brightness sensitivity test on the user to obtain the user's brightness sensitivity to the image; Conducting a color sensitivity test on the user to obtain the user's color sensitivity to the image; Conducting a contrast sensitivity test on the user to obtain the user's contrast sensitivity to the image; Conducting a texture complexity preference test on users to obtain their texture complexity preferences for images; The personalized perception configuration includes the brightness sensitivity, color sensitivity, contrast sensitivity and texture complexity preference.

3. The method according to claim 2, characterized in that The step of generating a perception weight map associated with the target image according to the personalized perception configuration includes: Generating a user perception vector according to the personalized perception configuration, the user perception vector comprising a brightness sensitivity dimension, a color sensitivity dimension, a contrast sensitivity dimension, and a texture complexity preference dimension; Generate a feature map of the target image, each pixel in the feature map includes brightness features, color features, contrast features and texture features; For the feature map, each pixel in the feature map is weightedly superimposed with the user perception vector to obtain the perception weight map.

4. The method according to claim 3, characterized in that The step of dividing the perception weight map into a plurality of regions and calculating the weight of each region includes: dividing the perceptual weight map into a plurality of regions; For each region, calculating the weighted average value of the pixels in the region; The weight average is used as the weight of the region.

5. The method according to claim 4, characterized in that Determining the importance level of the area according to the weight of each area includes: For each region, when the weight of the region is greater than a first threshold, determining the importance level of the region to be the first level; When the weight of the area is greater than a second threshold and less than or equal to the first threshold, determining the importance level of the area to be a second level; When the weight of the area is greater than a third threshold and less than or equal to the second threshold, determining the importance level of the area to be a third level; Among them, the third threshold is smaller than the second threshold, the second threshold is smaller than the first threshold, and the importance of the first level, the second level, and the third level decreases in sequence.

6. The method according to claim 5, characterized in that The compressing the portion of the target image associated with the region according to the importance level comprises: When the importance level of the region is a first level, compressing a portion of the target image associated with the region to a first degree; When the importance level of the region is a second level, compressing a portion of the target image associated with the region to a second degree; When the importance level of the region is the third level, compressing the portion of the target image associated with the region to a third degree; Among them, at the first degree, the second degree, and the third degree, the compression degree of the part of the target image associated with the area gradually increases.

7. The method according to claim 6, characterized in that The image is a JPEG image, The compressing the portion of the target image associated with the region to a first degree comprises: compressing the portion of the target image associated with the region by a quality factor of 10-30; The compressing the portion of the target image associated with the region to a second degree comprises: compressing the portion of the target image associated with the region by a quality factor of 50-70; The compressing the portion of the target image associated with the region to a third degree includes: compressing the portion of the target image associated with the region by a quality factor of 90-100.

8. A high-precision CMOS image sensor data storage optimization device based on an adaptive compression algorithm, characterized in that: include: The first module is used to determine the user's personalized perception configuration of the image based on the user's behavior data with the user's permission; The second module is used to obtain the target image collected by the high-precision CMOS image sensor; A third module is used to generate a perceptual weight map associated with the target image according to the personalized perceptual configuration; The fourth module is used to divide the perception weight map into multiple regions and calculate the weight of each region; A fifth module is used to determine the importance level of the area according to the weight of each area; The sixth module is used to compress the portion of the target image associated with the region according to the importance level.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

10. A device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 7.

Images