WO2022095742A1

WO2022095742A1 - Image and video processing methods and systems, and data processing device and medium

Info

Publication number: WO2022095742A1
Application number: PCT/CN2021/126137
Authority: WO
Inventors: 梁建华
Original assignee: 晶晨半导体(上海)股份有限公司
Priority date: 2020-11-05
Filing date: 2021-10-25
Publication date: 2022-05-12
Also published as: CN114449181A; CN114449181B

Abstract

Image and video processing methods and systems, and a data processing device and a medium. The image processing method comprises: extracting image features of an original image to obtain a first detailed image; selecting a target pixel and a local area from the first detailed image, the local area comprising the target pixel; calculating corresponding statistical feature information on the basis of a statistical relationship between other pixels in the local area and the target pixel, and updating the statistical feature information to color information of the target pixel in the first detailed image to obtain a second detailed image; and synthesizing the second detailed image and the original image to obtain a synthesized image. The scheme can be used to improve both of the image quality and the processing efficiency.

Description

Image and video processing method and system, data processing device and medium

technical field

The embodiments of this specification relate to the technical field of data processing, and in particular, to an image and video processing method and system, data processing device, and medium.

Background technique

With the development of consumer electronics, people can shoot and publish multimedia information such as videos and images anytime and anywhere. In addition, various multimedia information can be obtained through communication channels. It can be said that modern people live in the era of information explosion.

In order to improve the user's viewing experience, the display demand of multimedia information on different devices has also increased, which has prompted the rapid development of display technology, from CRT (Cathode Ray Tube, cathode ray tube) technology, LCD (Liquid Crystal Display, liquid crystal display) Technology to OLED (Organic Light-Emitting Diode, organic light-emitting diode, or Organic Electroluminescence Display, organic light-emitting semiconductor) technology, and Micro LED (a light-emitting diode miniaturization and matrix technology) and other technologies still under research, display technology More and more high-end, as a result, the display resolution is also getting higher and higher, the mainstream resolution from SD (Standard Definition, standard definition), HD (High Definition, high definition), FHD (Full High Definition, full HD) to 4K ( An ultra-high-definition display technology with a horizontal resolution at or near 4096 pixels), and higher-end display resolutions such as 8K (an ultra-high-definition display technology with a horizontal resolution at or near 8192 pixels) have also emerged.

However, due to the different shooting techniques in different periods, combined with the influence of environmental factors, human factors, etc., the imaging quality of multimedia information shot in the past is poor, such as there is more noise, and the pixel resolution is low. The display device has a higher display resolution, and the pixel levels of the two do not match. Therefore, when displaying multimedia information taken in the past on the existing display device, problems such as unclear picture and rough picture quality are prone to occur, affecting users. viewing experience. In order to solve the above problems, image enhancement technology can be used to optimize the multimedia information recorded in the past.

The image enhancement technology may specifically include a resolution enhancement method, a frame rate enhancement method, a pixel quality enhancement method, and the like. Among them, the resolution enhancement method is more commonly used in electronic products such as mobile phones, computers, and televisions. The resolution enhancement method amplifies the multimedia information of low resolution (LR), hoping to achieve an ideal high resolution (High Resolution, HR) state, enhance the resolution of multimedia information, and then increase the resolution of multimedia information at the pixel level. Better viewing experience on taller display devices.

The traditional resolution enhancement methods have problems such as blurring and confusion. In order to solve the problems existing in the traditional resolution enhancement methods, an improved resolution enhancement method and a deep learning-based resolution enhancement method are proposed. However, although the improved resolution enhancement method optimizes the edge of the image, the algorithm logic is more complex and does not optimize the rest of the image; while the resolution enhancement method based on deep learning can improve the overall image quality, it consumes computing resources huge, and the implementation cost is high.

To sum up, it can be seen that there are still many problems in the existing resolution enhancement methods, which need to be improved by technicians.

technical problem

In view of this, the embodiments of the present specification provide an image and video processing method, system, data processing device, and medium thereof, which can improve both image quality and processing efficiency.

technical solutions

An embodiment of this specification provides an image processing method, including: extracting image features of an original image to obtain a first detail image; selecting target pixels and a local area from the first detail image, where the local area includes the target pixel; based on the statistical relationship between other pixels in the local area and the target pixel, calculate the corresponding statistical feature information, and update the statistical feature information to the color of the target pixel in the first detail image information to obtain a second detailed image; and combining the second detailed image and the original image to obtain a composite image.

The embodiments of the present specification also provide a video processing method, including: extracting image features of a target video frame in a video stream to obtain a first detail image; selecting target pixels and local areas from the first detail image, the local The target pixel is included in the area; based on the statistical relationship between other pixels in the local area and the target pixel, the corresponding statistical feature information is calculated, and the statistical feature information is updated to the first detail image The color information of the target pixel is obtained to obtain a second detail image; the second detail image and the target video frame are combined to obtain a composite video frame.

The embodiments of this specification also provide an image processing system, including: a detail extraction module, adapted to extract image features of an original image to obtain a first detail image; a detail generation module, adapted to select a target from the first detail image pixel and local area, the local area includes the target pixel, and based on the statistical relationship between other pixels in the local area and the target pixel, the corresponding statistical feature information is calculated, and the statistical feature information is calculated. The color information of the target pixel in the first detailed image is updated to obtain a second detailed image; the image synthesis module is adapted to combine the second detailed image and the original image to obtain a composite image.

The embodiments of this specification also provide a video processing system, including: a detail extraction module, adapted to extract image features of a target video frame in a video stream to obtain a first detail image; a detail generation module, adapted to extract the first detail image from the first detail image The target pixel and local area are selected in the image, the local area includes the target pixel, and based on the statistical relationship between other pixels in the local area and the target pixel, the corresponding statistical feature information is calculated and obtained. The statistical feature information is updated to the color information of the target pixel in the first detailed image to obtain a second detailed image; a video synthesis module is adapted to combine the second detailed image and the target video frame to obtain a composite video frame.

Embodiments of the present specification further provide a data processing device, including a memory and a processor, where the memory stores computer instructions that can be run on the processor, and the processor executes any of the above when running the computer instructions. The steps of the method of an embodiment.

The embodiments of the present specification further provide a computer-readable storage medium on which computer instructions are stored, and when the computer instructions are executed, the steps of the method described in any one of the foregoing embodiments are executed.

beneficial effect

Using the image processing solution provided in this specification, after the image features of the original image are extracted and the first detailed image is obtained, the corresponding statistical feature information can be obtained by calculating the statistical relationship between the target pixel and other pixels in the local area, Thereby, the statistical characteristics between the image pixels are used to predict the subtle image features implicit in the local area, and after the statistical feature information is updated to the color information of the corresponding target pixel, the obtained second detailed image contains the subtle image features, Then, the second detailed image and the original image are combined to enrich the image features of the original image, reasonably improve the defect of blurred details of the original image, make the image have stronger detail expression ability, and improve the image quality. The logic complexity of the image processing solution of the embodiment is low, and the processing efficiency is improved. Therefore, the image processing solution provided in this specification can improve both the image quality and the processing efficiency.

Using the video processing solution of the embodiment of this specification, after the image features of the target video frame in the video stream are extracted to obtain the first detailed image, the corresponding Statistical feature information, so as to use the statistical properties between image pixels to predict the subtle image features implicit in the local area, after updating the statistical feature information to the color information of the corresponding target pixel, the obtained second detail image contains Subtle image features, and then the second detailed image and the target video frame are combined to enrich the image features of the target video frame, reasonably improve the defect of blurred details of the video picture, so that the image has a stronger ability to express details and improve the performance of the target video frame. image quality, and the logic complexity of the video processing solution in the embodiments of this specification is low, which improves processing efficiency, so the video processing solution provided in this specification can improve both image quality and processing efficiency.

Description of drawings

In order to explain the technical solutions of the embodiments of the present specification more clearly, the following briefly introduces the drawings that are required to be used in the embodiments of the present specification or the description of the prior art. Obviously, the drawings described below are only for the purposes of the present specification. In some embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.

FIG. 1 is a flowchart of an image processing method in an embodiment of the present specification.

FIG. 2 is a schematic diagram of a filtering method in an embodiment of the present specification.

Fig. 3a is a schematic diagram of an original image in the embodiment of the present specification.

FIG. 3b is a schematic diagram of a first detailed image in an embodiment of the present specification.

FIG. 4 is a schematic diagram of selecting a target pixel and a local area in an embodiment of the present specification.

FIG. 5 is a schematic diagram of another selection of target pixels and local regions in the embodiment of the present specification.

FIG. 6 is a schematic diagram of a partial region in the embodiment of the present specification.

FIG. 7 is a schematic diagram of a partial area of a target pixel after the color information of the target pixel is updated according to an embodiment of the present specification.

FIG. 8 is a schematic diagram of a composite image in an embodiment of the present specification.

FIG. 9 is a structural block diagram of an image processing system in an embodiment of the present specification.

FIG. 10 is a flowchart of an image processing system in the embodiment of the present specification.

FIG. 11 is a flowchart of a video processing method in an embodiment of the present specification.

FIG. 12 is a structural block diagram of a video processing system in an embodiment of this specification.

Embodiments of the present invention

As described in the background art, the resolution enhancement method is more commonly used in electronic products such as mobile phones, computers, and televisions. In practical applications, traditional resolution enhancement methods may include: a nearest neighbor interpolation (Nearest interpolation) algorithm, a bilinear interpolation (Bilinear interpolation) algorithm, a bicubic interpolation (Bicubic interpolation) algorithm, and the like. However, the images obtained by the above-mentioned traditional resolution enhancement methods are prone to blurring and aliasing problems.

After the research of technicians, some improved resolution enhancement methods are proposed, such as Directional Interpolation algorithm, Example based super resolution algorithm, Sparse representation based super resolution algorithm. super resolution) algorithm, etc. These improved resolution enhancement methods can obtain better results than the traditional resolution enhancement methods, but the algorithm logic is more complex, and these improved resolution enhancement methods only optimize the edge part of the image, for the rest of the image Parts have not been improved.

In addition, due to the rise of deep learning, researchers have also proposed some resolution enhancement methods based on deep learning, such as deep learning based super resolution algorithms. Thanks to the powerful nonlinear expression ability of deep neural network, the resolution enhancement method based on deep learning can improve the overall quality of the image. For example, neural networks such as GAN (Generative Adversarial Network, Generative Adversarial Network) can be used to implement deep learning to add virtual details to images. However, the algorithmic logic of the resolution enhancement method based on deep learning is more complicated, and a large amount of data calculation is required, which consumes a lot of computing resources.

In order to solve the technical problems existing in the existing resolution enhancement methods, the embodiments of this specification provide an image processing solution. After the image features of the original image are extracted, the statistical characteristics between the pixels in the local area can be used to update the image features in the local area. The color information of pixels, thereby increasing subtle image features, improving image quality, reducing the complexity of algorithm logic, and improving processing efficiency.

In order for those skilled in the art to more clearly understand and implement the concepts, implementation solutions and advantages of the embodiments of the present specification, the following detailed description is given through specific application scenarios with reference to the accompanying drawings.

Referring to the flowchart of an image processing method in the embodiment of the present specification shown in FIG. 1, in the embodiment of the present specification, the method may include the following steps:

SA1, extract the image features of the original image to obtain the first detail image.

The image features may include: color features, texture features, shape features, spatial relationship features, and the like. The original image may be a color image, a grayscale image, or a black and white image. In addition, the original image can be obtained by reading the image data recorded on the designated storage address.

Optionally, before extracting the image features of the original image, the original image may be grayscaled to obtain a grayscale original image, which is used as the object of feature extraction processing; or, before extracting the image features of the original image, you can The original image is binarized to obtain a black and white original image, which is used as the object of feature extraction.

In specific implementation, according to image requirements and actual application scenarios, corresponding feature extraction methods can be used to extract image features such as color features, texture features, shape features, and spatial relationship features of the original image, thereby obtaining detailed information carried by the original image. , to obtain the first detailed image, which is used as the basis for the subsequent acquisition of the second detailed image.

The following describes in detail how to acquire the first detailed image through specific embodiments.

In an optional example, an image feature of the original image can be extracted by using a filtering method on the original image. Specifically, as shown in FIG. 2 , based on the preset geometric parameters of the filter window, the corresponding filter window Blk is generated. In the original image P1, the filter window Blk is in accordance with the preset moving direction and the preset moving step size, and Based on the preset filter coefficients, filter processing is performed on the pixels in the filter window Blk to extract image features of the original image. The filtering processing may include mean filtering processing, median filtering processing, Gaussian filtering processing, bilateral filtering processing, etc., and the filtering processing may be performed based on a two-dimensional coordinate system.

As shown in FIG. 2 , a rectangular frame formed by the dotted line in the original image P1 represents one pixel, the moving direction can be set in the order from left to right and from top to bottom, and the preset moving step can be one pixel . The first detail image is obtained after the filtering process of the filtering window Blk.

It can be understood that, for the convenience of description, the original image shown in FIG. 2 does not contain any image features. In practical applications, the original image may contain more pixels and have more abundant image features. This does not limit. For example, as shown in Fig. 3a, which is a schematic diagram of an original image, the first detail image P3 can be obtained after the original image P2 is filtered. Compared with the original image P2 shown in Fig. 3a, the first detail image P3 shown in Fig. 3b A detail image P3 removes the pixel information that changes gently, and retains the pixel information of the area that changes significantly.

It can also be understood that FIG. 2 is only a schematic illustration of using the filtering method to obtain the first detailed image. When actually implementing the technical solution of this specification, the relevant parameters of the filtering method can be adjusted according to the filtering requirements. For example, the filtering window can be adjusted. At least one parameter among the geometric parameters and the filter coefficients of the filtering process, so as to achieve the required filtering effect and obtain the first detailed image that meets the filtering requirements. The embodiments of the present specification do not limit this.

It should be noted that the above embodiments are only illustrative, and do not limit the embodiments of the present specification to extract image features through other methods. For example, the local binary pattern (LBP) method can also be used to extract image features.

SA2: Select a target pixel and a local area from the first detail image, and the local area includes the target pixel.

The target pixel and the local area may be selected according to actual settings. For example, the selection method of the target pixel and the local area may include:

Based on a preset selection condition, the target pixel is selected from the first detail image, and a local area including the target pixel is selected based on a preset area geometric parameter.

In specific implementation, the selection conditions and regional geometric parameters can be set according to actual needs and scenarios, for example, target pixels can be selected according to a preset position, or target pixels can be selected according to a preset order; for example, the regional geometric parameters It can include parameters of regular graphics such as circles, rings, and lines, as well as parameters of irregular graphics. The embodiments of the present specification do not limit this.

In an embodiment of the present invention, the selection condition may be: select the pixel in the 3×i row and the 3×j column as the target pixel, and the regional geometric parameters may include: a side length of 5×5 (pixels) , where i and j are both positive integers. As shown in FIG. 4 , taking i=1 and j=1 as an example, in the first detailed image P4, the pixel A in the third row and the third column is the target pixel, and a 5×5 (pixel) size is generated. Rectangular local area so that the resulting local area contains pixel A. By analogy, at least one eligible target pixel and its corresponding local area can be obtained.

In another embodiment of the present invention, the selection condition may be: select the pixel in the 3×i row and the 3×j column as the target pixel, and the area geometric parameter may be a 5×5 (pixel) cross shape . . As shown in FIG. 5 , taking i=1 and j=1 as an example, in the first detail image P5, the pixel A in the third row and the third column is the target pixel, and a 5×5 (pixel) size is generated. A cross-shaped local area, so that the resulting local area contains pixel A. By analogy, at least one eligible target pixel and its corresponding local area can be obtained.

In a further achievable example, in order to facilitate the generation of the local area, the local area may be generated with the target pixel as the center pixel. For example, with continued reference to FIG. 4 , a local area Fa of a size of 5×5 (pixels) may be generated with the pixel A as the center. For another example, continuing to refer to FIG. 5 , a local area Fb with a size of 5×5 (pixels) may be generated with the pixel A as the center.

It can be understood that, for the convenience of description, the first detail images shown in FIG. 4 and FIG. 5 do not contain any image features. In practical applications, the first detail images may contain more pixels and have more abundant images. feature, which is not limited in the embodiments of this specification.

SA3, based on the statistical relationship between other pixels in the local area and the target pixel, calculate the corresponding statistical feature information, and update the statistical feature information to the color information of the target pixel in the first detail image , to get the second detail image.

Wherein, the statistical relationship can represent the relationship between other pixels in the local area and the relevant statistics of the target pixel; the statistical feature information can be regarded as a kind of statistics, and the statistical feature information can be used for Characterize the subtle image features implicit in the local area.

In the specific implementation, based on the first detail image, the selected local area can represent the local native detail in the first detail image, and statistical feature information can be obtained through the statistical characteristics between the pixels in the local area, and then the target pixel can be updated. to obtain a second detailed image containing subtle image features.

In an embodiment of the present specification, as shown in FIG. 6 , the local area Fc is a rectangular area with a size of 5×5 (pixels) selected from the first detail image, and the local area Fc is divided by a dotted line and formed The rectangular box represents a pixel, and the target pixel C is the central pixel of the local area Fc. Based on the statistical relationship between the other 24 pixels in the local area Fc and the target pixel C, the corresponding statistical feature information is calculated, and the The statistical feature information is updated to the color information of the target pixel in the first detailed image to obtain a second detailed image.

For ease of understanding, Fig. 7 only shows the local area Fc' after the color information of the corresponding target pixel in the second detailed image is updated. Compared with Fig. 6, it can be seen that the color information of the target pixel C has changed.

In practical applications, different statistics can be obtained according to different statistical relationships, such as variance value, standard deviation value, mean value, weighted value, median value, etc. However, in the embodiments of the specification, if the median or mean value is used As a statistic, that is, the median or average value between other pixels in the local area and the target pixel is calculated, the high-frequency band pixel information is suppressed, and the obtained statistical feature information is always located in the color information interval of the pixels in the local area, and is lower than the color information of some high-frequency pixels. Therefore, when the statistic is the mean or median value, the process of generating statistical feature information can be regarded as a low-pass filtering, thereby reducing the number of image features in the high-frequency band in the local area. Causes image details to be blurred and new details unpredictable.

In order to avoid the above situation, when calculating the corresponding statistical feature information based on the statistical relationship between other pixels in the local area and the target pixel, a statistical relationship that pays more attention to high-frequency pixel information, such as variance, standard difference, weight, etc.

In a specific implementation, in order to reduce the amount of calculation and ensure the color invariance and rotation invariance of the local area, a color condition can be set. After acquiring the color information of other pixels in the local area, you can For color conditions, the pixels that meet the color conditions are determined from the local area, and the calculation is performed based on other pixels that meet the color conditions in the local area to obtain the corresponding statistical feature information of the target pixels.

Optionally, the color condition is set according to original color information of the target pixel. For example, if the original color information of the target pixel is pixel1, then pixel1 can be used as the color threshold, and when the color information of other pixels is not less than the color threshold pixel1, the other pixels meet the color condition.

Further, the calculation based on other pixels that meet the color conditions in the local area to obtain the corresponding statistical feature information of the target pixel may specifically include: obtaining a preset statistical coefficient matrix, the statistical coefficient matrix and the statistical coefficient matrix. There is a corresponding relationship between the pixels in the local area. According to the position information of other pixels in the local area that meet the color conditions, the statistical coefficients of the corresponding positions are obtained from the statistical coefficient matrix, and the statistical features are calculated according to the preset statistical relationship. information.

In a specific implementation, when there are multiple corresponding target pixels and local areas, the corresponding target pixels and local areas may be processed according to a preset order, or at least part of the corresponding target pixels may be obtained according to a preset number and local areas are processed. This specification does not limit the processing order of the plurality of corresponding target pixels and local regions.

SA4, combine the second detail image and the original image to obtain a composite image.

In a specific implementation, as shown in FIG. 8 , the second detailed image P6 may be superimposed with the original image P2 to obtain a composite image P2'.

It can be seen from the above that, by using the image processing method provided in this specification, the corresponding statistical feature information can be calculated through the statistical relationship between the target pixel and other pixels in the local area, so that the statistical features between image pixels can be used to predict. The subtle image features hidden in the local area, after the statistical feature information is updated to the color information of the corresponding target pixel, the obtained second detailed image contains subtle image features, and then the second detailed image and the The original images are merged to enrich the image features of the original images, and reasonably improve the defects of blurred details of the original images, so that the images have a stronger ability to express details and improve the image quality.

Compared with the above-mentioned traditional and improved resolution enhancement methods such as bicubic interpolation and direction-based interpolation, the image processing method provided in this specification can reasonably improve the defect of blurred details of the original image; compared with the resolution enhancement method of deep learning, The image processing method provided in this specification greatly reduces the complexity of algorithm logic and improves processing efficiency while effectively predicting the subtle image features implicit in the local area.

In conclusion, the image processing method provided in this specification can improve both image quality and processing efficiency.

In practical applications, since the image processing method provided in this specification improves processing efficiency, it can meet the needs of real-time image processing, enable existing display devices to display composite images with more details in real time, optimize visual effects, and effectively Improve user viewing experience.

In addition, since the image processing method provided in this specification can improve both image quality and processing efficiency, it has a wider range of application scenarios. It can not only improve the image quality of the images captured in the past, but also can be used when the resolution of the shooting device is insufficient. , the quality of the image data collected by the photographing device is improved by the image processing method provided in this specification, so that the image displayed in real time on the display device has more details.

In a specific implementation, since more attention is paid to the high-frequency band pixel information when acquiring statistical feature information, the pixels in the second detailed image may be in a higher frequency band, which is likely to generate noise. Therefore, when combining the second detailed image and the Before the original image, noise reduction processing may be performed on the second detail image, so as to combine the second detail image after noise reduction processing with the original image.

The following is a schematic illustration of how to perform noise reduction processing on the second detail image through several embodiments.

In an embodiment of the present specification, before merging the second detail image and the original image, smoothing may be performed on the second detail image, so as to combine the smoothed second detail image and the original image Merge. Wherein, the smoothing process may select a corresponding filtering method according to the actual situation, such as a Gaussian low-pass filtering method, a bilateral filtering method, and the like. The embodiments of the present specification do not limit this.

Further, the filtering method with fixed parameters can remove the noise in the image, but it will lose a lot of image details, making the image blurred. detail. For example, the smoothing processing on the second detail image may include: performing guide filtering processing on the second detail image based on a preset guide image.

Furthermore, in order to make the intensity of the guided filtering process (ie, the value range of the color information) adaptive according to different second detail images, the original image can be used as the guide image, and based on the original image, The second detail image is subjected to guided filtering processing.

In another embodiment of the present specification, before merging the second detail image and the original image, image modulation processing may be performed on the second detail image based on a preset modulation image, so that the modulated image The second detail image and the original image are merged. Wherein, the modulated image may be an image set according to modulated prior data or modulated test data.

In specific implementation, the image processing method provided in this specification can be combined with the method for realizing image enhancement, and has better compatibility and adaptability.

For example, before extracting the image features of the original image, the method may further include: using a resolution boosting method to perform a resolution enlarging process on the original image to extract the image features of the original image after the resolution enlarging process, and The second detailed image and the original image after resolution enlargement processing are combined. Therefore, through the resolution amplification process, the resolution of the original image can be improved, which is beneficial to the processing in the subsequent steps.

It can be understood that the above describes the multiple embodiments provided by the embodiments of this specification, and the optional modes introduced by the embodiments can be combined and cross-referenced with each other without conflict, thereby extending a variety of possibilities. These can be considered as the embodiments disclosed and disclosed in this specification.

The present specification also provides an image processing system corresponding to the above-mentioned image processing method, which will be described in detail below through specific embodiments with reference to the accompanying drawings. It should be known that the image processing system described below can be considered as the functional modules required to implement the image processing method provided in this specification; the content of the image processing system described below can be compared with the content of the image processing method described above. refer to each other.

Referring to the structural block diagram of an image processing system in the embodiment of the present specification shown in FIG. 9, in the embodiment of the present specification, the image processing system 90 may include: a detail extraction module 91, adapted to extract image features of the original image, A first detail image is obtained; the detail generation module 92 is adapted to select a target pixel and a local area from the first detail image, the local area includes the target pixel, and based on other pixels in the local area and all According to the statistical relationship between the target pixels, the corresponding statistical feature information is obtained by calculation, and the statistical feature information is updated to the color information of the target pixel in the first detailed image to obtain a second detailed image; the image synthesis module 93, suitable for and combining the second detail image and the original image to obtain a composite image.

It can be seen from the above that using the image processing system provided in this specification, through the statistical relationship between the target pixel and other pixels in the local area, the corresponding statistical feature information can be calculated, so that the statistical features between image pixels can be used to predict. The subtle image features hidden in the local area, after the statistical feature information is updated to the color information of the corresponding target pixel, the obtained second detailed image contains subtle image features, and then the second detailed image and the The original images are merged to enrich the image features of the original images, and reasonably improve the defects of blurred details of the original images, so that the images have a stronger ability to express details and improve the image quality.

Compared with the above-mentioned systems using traditional and improved resolution enhancement methods such as bicubic interpolation and direction-based interpolation, the image processing system provided in this specification can reasonably improve the defect of blurred details of the original image; The image processing system provided in this specification can effectively predict the subtle image features implicit in the local area, and at the same time, it can also greatly reduce the complexity of the algorithm logic and improve the processing efficiency.

In addition, since the image processing system provided in this specification can take into account the improvement of image quality and processing efficiency, it has a wider range of application scenarios. , through the image processing system provided in this specification, the quality of the image data collected by the photographing device is improved, so that the image displayed in real time on the display device has more details.

In a specific implementation, the detail extraction module can obtain the original image by reading the image data recorded on the specified storage address, wherein the detail extraction module can adopt the reading method of direct reading, or indirectly through other modules. The reading method of reading is not limited in the embodiment of this specification.

Moreover, according to the image requirements and actual application scenarios, the detail extraction module can use the corresponding feature extraction method to extract image features such as color features, texture features, shape features, and spatial relationship features of the original image, thereby obtaining the detail information carried by the original image. , to obtain the first detailed image, which is used as the basis for the subsequent acquisition of the second detailed image. For the specific implementation process of the detail extraction module, reference may be made to the content description of the relevant part of the image processing method, which will not be repeated here.

In a specific implementation, the detail generation module may select the target pixel and the local area according to actual settings. For example, the detail generation module may select the target pixel and the local area in the following selection manner: select the target pixel from the first detail image based on a preset selection condition, and select the target pixel from the first detail image based on a preset selection condition. Geometry parameters, select the local area containing the target pixel.

Then, the detail generation module may use the first detail image as a basis, the selected local area may represent the local native detail in the first detail image, and statistical feature information can be obtained through statistical characteristics between pixels in the local area, and then Update the color information of the target pixel to obtain a second detailed image containing subtle image features. For the specific implementation process of the detail generation module, reference may be made to the content description of the relevant part of the image processing method, which will not be repeated here.

In a specific implementation, when there are multiple corresponding target pixels and local areas, the detail generation module may process the corresponding target pixels and local areas in a preset order, or obtain at least Part of the corresponding target pixels and local regions are processed. This specification does not limit the order in which the detail generation module processes the plurality of corresponding target pixels and local regions.

In the specific implementation, since more attention is paid to the high-frequency band pixel information when acquiring statistical feature information, the pixels in the second detailed image may be in a higher frequency band, which is prone to generate noise. In order to reduce the noise of the second detailed image, as shown in Fig. As shown in FIG. 9 , the image processing system 90 may further include: a detail processing module 94 .

The detail processing module 94 is located between the detail generation module 92 and the image synthesis module 93, and is adapted to perform noise reduction processing on the second detail image obtained by the detail generation module 92, and combine the noise reduction processed image. The second detail image is sent to the image synthesis module 93 . The image synthesis module 93 combines the second detail image after noise reduction processing and the original image to obtain a composite image.

It can be understood that, for the specific implementation process of the detail processing module, reference may be made to the content description of the relevant part of the image processing method, which will not be repeated here.

In a specific implementation, the image processing system provided in this specification can process an image that has undergone image enhancement, and has better compatibility and adaptability.

For example, as shown in FIG. 9 , the image processing system 90 may further include: a resolution processing module 95 . The resolution processing module 95 is respectively connected with the detail extraction module 91 and the image synthesis module 93, and is adapted to perform resolution enlargement processing on the original image, and send the original image after the resolution enlargement processing to the The detail extraction module 91 and the image synthesis module 93 are described. The detail extraction module 91 extracts the image features of the original image after resolution enlargement processing to obtain a first detail image; the image synthesis module 93 combines the second detail image and the original image after resolution enlargement processing to obtain a synthesis image.

Therefore, through the resolution amplification process, the resolution of the original image can be improved, which is beneficial to the processing of subsequent modules.

In a specific implementation, the resolution processing module can obtain the original image by reading the image data recorded on the specified storage address, wherein the resolution processing module can adopt the reading method of direct reading, or can adopt other methods of reading. The reading method of the module indirect reading is not limited in the embodiment of this specification.

In order to facilitate those skilled in the art to understand the workflow of the image processing system provided in this specification, a schematic description is given below with reference to the flowchart of the image processing system.

As shown in FIG. 10, which is a flowchart of an image processing system, the image processing system 10 may include a resolution processing module 101, a detail extraction module 102, a detail generation module 103, a detail processing module 104, and an image synthesis module 105. The workflow of the image processing system 10 is as follows.

1) The original image Img0 is input to the resolution processing module 101, and the resolution processing module 101 performs resolution enlargement processing on the original image Img0, and outputs the original image Img0' after the resolution enlargement processing to the details Extraction module 102 and the image synthesis module 105 .

2) The detail extraction module 102 extracts the image features of the original image Img0' after the resolution amplification process, and outputs the first detail image Img1 to the detail generation module 103.

3) The detail generation module 103 selects a target pixel and a local area from the first detail image Img1, the local area includes the target pixel, and based on the relationship between other pixels in the local area and the target pixel The statistical relationship between the two is calculated to obtain the corresponding statistical feature information, the statistical feature information is updated to the color information of the target pixel in the first detail image, and the second detail image Img2 is output to the detail processing module 104.

4) The detail processing module 104 performs noise reduction processing on the second detail image Img2, and outputs the second detail image Img2' after the noise reduction processing to the image synthesis module 105.

5) The image synthesis module 105 merges the second detail image Img2' after the noise reduction processing and the original image Img0' after the corresponding resolution amplification processing, and outputs the synthesized image ImgH.

It can be understood that the above describes the multiple embodiments provided by the embodiments of the present invention, and the optional modes introduced by the embodiments can be combined and cross-referenced with each other without conflict, thereby extending a variety of possibilities. These can be considered as the embodiments disclosed and disclosed in the present invention.

According to the above content, the existing resolution enhancement methods have problems such as ambiguity, confusion, and overly complex algorithm logic, which makes it more difficult to apply to video processing with a larger amount of image data. In order to solve the technical problems existing in the existing resolution enhancement methods, The embodiments of this specification also provide a video processing solution. After extracting the image features of the target video frame, the color information of the pixels in the local area can be updated by using the statistical characteristics between the pixels in the local area, thereby increasing the target video frame. It can improve the image quality, reduce the complexity of algorithm logic and improve processing efficiency.

Referring to the flowchart of a video processing method in the embodiment of the present specification shown in FIG. 11 , in the embodiment of the present specification, the method may include the following steps.

SB1, extract the image features of the target video frame in the video stream to obtain the first detail image.

In a specific implementation, the video stream may be an encoded and compressed video stream (also referred to as an encoded stream), or may be an unencoded and compressed video stream (also referred to as an original stream). If the video stream has been encoded and compressed, before extracting the image features of the target video frame in the video stream, the video stream may be reversely decoded to obtain the original stream.

A video frame included in the video stream can be regarded as an image, and by continuously playing the video frames, a dynamic picture can be displayed on the display device. Therefore, after the target video frame in the video stream is determined through the preset target selection condition, image feature extraction can be performed on the target video frame.

Optionally, before extracting the image features of the target video frame, the target video frame can be subjected to grayscale processing to obtain a grayscale target video frame as the object of feature extraction processing; or, after extracting the image of the target video frame Before the feature, the target video frame can be binarized to obtain a black and white target video frame, which can be used as the object of feature extraction processing.

It can be understood that different target selection conditions can be set according to actual needs and application scenarios, such as selecting video frames according to time intervals, or selecting video frames with specified serial numbers according to sorting, etc.; and, according to target selection conditions, one or more target video frames, the embodiments of this specification do not limit the specific content of the target selection condition and the number of target video frames.

In specific implementation, according to image requirements and actual application scenarios, corresponding feature extraction methods can be used to extract image features such as color features, texture features, shape features, and spatial relationship features of the target video frame, thereby obtaining the image features carried by the target video frame. The detail information is obtained, and the first detail image is obtained, which is used as the basis for the subsequent acquisition of the second detail image. For example, filtering processing may be performed on the target video frame based on preset geometric parameters of the filtering window and filter coefficients, and the image features of the target video frame may be extracted.

It can be understood that, for the specific implementation process of performing image feature extraction on the target video frame, reference may be made to the content description of the relevant part of the above-mentioned image processing method, which will not be repeated here.

SB2: Select a target pixel and a local area from the first detail image, and the local area includes the target pixel.

In a specific implementation, the target pixel and the local area may be selected according to actual settings. For example, the selection method of the target pixel and the local area may include: selecting the target pixel from the first detail image based on a preset selection condition, and selecting the target pixel including the describe the local area of the target pixel.

For the specific implementation process of selecting the target pixel and the local area, reference may be made to the content description of the relevant part of the above-mentioned image processing method, which will not be repeated here.

SB3, based on the statistical relationship between other pixels in the local area and the target pixel, calculate the corresponding statistical feature information, and update the statistical feature information to the color information of the target pixel in the first detail image , to get the second detail image.

Further, in order to reduce the amount of calculation and ensure the color invariance and rotation invariance of the local area, a color condition can be set. After acquiring the color information of other pixels in the local area, the preset color condition can be used. , determine the pixels that meet the color conditions from the local area, and perform calculations based on other pixels that meet the color conditions in the local area to obtain the corresponding statistical feature information of the target pixel. Wherein, the color condition is set according to the original color information of the target pixel.

It can be understood that, for the specific implementation process of obtaining the second detailed image, reference may be made to the content description of the relevant part of the above-mentioned image processing method, which will not be repeated here.

SB4: Combine the second detailed image and the target video frame to obtain a composite video frame.

In specific implementation, the second detail image may be directly merged with the corresponding target video frame in the video stream, or the target video frame may be extracted from the video stream, and the second detail image may be combined with the second detail image. Insert into the video stream after merging.

It can be seen from the above that by using the video processing method provided in this specification, through the statistical relationship between the target pixel and other pixels in the local area, the corresponding statistical feature information can be calculated, so that the statistical features between image pixels can be used to predict. The subtle image features hidden in the local area, after the statistical feature information is updated to the color information of the corresponding target pixel, the obtained second detailed image contains subtle image features, and then the second detailed image and the The target video frames are merged to enrich the image features of the target video frames, reasonably improve the defect of blurred details of the target video frames, so that the target video frames have a stronger ability to express details and improve the image quality.

Compared with the above-mentioned traditional and improved resolution enhancement methods such as bicubic interpolation and direction-based interpolation, the video processing method provided in this specification can reasonably improve the defect of blurred details of target video frames; compared with the resolution enhancement method of deep learning , the video processing method provided in this specification greatly reduces the complexity of algorithm logic and improves the processing efficiency while effectively predicting the subtle image features implicit in the local area.

In conclusion, the video processing method provided in this specification can improve both image quality and processing efficiency.

In practical applications, since the video processing method provided in this specification improves processing efficiency, it can meet the needs of real-time video processing, enable existing display devices to display composite video frames with more details in real time, and optimize visual effects. Effectively improve user viewing experience.

In addition, since the video processing method provided in this specification can improve both image quality and processing efficiency, it has a wider range of application scenarios, which can not only improve the imaging quality of videos shot in the past, but also can be used when the resolution of the shooting equipment is insufficient. , the quality of the video frame data collected by the shooting device is improved by the video processing method provided in this specification, so that the video displayed in real time on the display device has more details.

In a specific implementation, if more attention is paid to high-frequency pixel information when acquiring statistical feature information, the pixels in the second detailed image may be in a higher frequency band, which is likely to generate noise. Therefore, when combining the second detailed image and the target Before the video frame, noise reduction processing may be performed on the second detail image, so as to combine the second detail image after noise reduction processing with the target video frame.

In an embodiment of this specification, before merging the second detail image and the target video frame, smoothing may be performed on the second detail image, so as to combine the smoothed second detail image and the target Video frames are merged. Wherein, the smoothing process may select a corresponding filtering method according to the actual situation, such as a Gaussian low-pass filtering method, a bilateral filtering method, and the like. The embodiments of the present specification do not limit this.

Further, in order to enable the intensity of the guided filtering process (that is, the value range of the color information) to be adaptive according to different second detail images, the target video frame can be used as a guide image, and based on the target video frame, A guided filtering process is performed on the second detail image.

In another embodiment of the present specification, before merging the second detail image and the target video frame, image modulation processing may be performed on the second detail image based on a preset modulation image, so as to The second detail image and the target video frame are merged. The modulated image may be an image set according to modulation prior data or modulation test data, or may be a picture selected from the video stream according to prior experience for distinguishing noise and details.

In specific implementation, the video processing method provided in this specification can be combined with the method for realizing image enhancement, and has good compatibility and adaptability.

For example, before extracting the image features of the target video frame, the method may further include: using a resolution enhancement method to perform a resolution enlargement process on the target video frame, so as to extract the image features of the target video frame after the resolution enlargement processing. , and combine the second detailed image and the target video frame after resolution enlargement processing. Therefore, through the resolution amplification process, the resolution of the original image can be improved, which is beneficial to the processing in the subsequent steps.

This specification also provides a video processing system corresponding to the above-mentioned video processing method, which will be described in detail below through specific embodiments with reference to the accompanying drawings. It should be known that the video processing system described below can be considered as the functional modules required to implement the video processing method provided in this specification; the content of the video processing system described below can be compared with the content of the video processing method described above. refer to each other.

Referring to the structural block diagram of a video processing system in the embodiment of the present specification shown in FIG. 12, in the embodiment of the present specification, the video processing system 120 may include: a detail extraction module 121, adapted to extract target video frames in a video stream The image features of the first detail image are obtained to obtain a first detail image; the detail generation module 122 is adapted to select a target pixel and a local area from the first detail image, the local area includes the target pixel, and based on the local area Calculate the statistical relationship between other pixels and the target pixel, calculate the corresponding statistical feature information, update the statistical feature information to the color information of the target pixel in the first detailed image, and obtain a second detailed image; video synthesis; The module 123 is adapted to combine the second detailed image and the target video frame to obtain a composite video frame.

It can be seen from the above that using the video processing system provided in this specification, through the statistical relationship between the target pixel and other pixels in the local area, the corresponding statistical feature information can be calculated, so that the statistical features between image pixels can be used to predict. The subtle image features hidden in the local area, after the statistical feature information is updated to the color information of the corresponding target pixel, the obtained second detailed image contains subtle image features, and then the second detailed image and the The target video frames are merged to enrich the image features of the target video frames, reasonably improve the defect of blurred details of the target video frames, so that the target video frames have a stronger ability to express details and improve the image quality.

Compared with the above-mentioned systems using traditional and improved resolution enhancement methods such as bicubic interpolation and direction-based interpolation, the video processing system provided in this specification can reasonably improve the defect of blurred details of target video frames; The system of the resolution enhancement method provided by this specification, while effectively predicting the subtle image features implicit in the local area, greatly reduces the algorithm logic and improves the processing efficiency.

In conclusion, the video processing system provided in this specification can improve both image quality and processing efficiency.

In practical applications, since the video processing system provided in this specification improves processing efficiency, it can meet the needs of real-time video processing, enable existing display devices to display synthetic video frames with more details in real time, and optimize visual effects. Effectively improve user viewing experience.

In addition, since the video processing system provided in this specification can improve both image quality and processing efficiency, it has a wider range of application scenarios, which can not only improve the image quality of videos shot in the past, but also can be used when the resolution of the shooting equipment is insufficient. , the quality of the video frame data collected by the shooting device is improved by the video processing system provided in this specification, so that the video displayed in real time on the display device has more details.

In a specific implementation, the detail extraction module can obtain the target video frame by reading the video frame data recorded on the designated storage address, or can read the video stream through the designated communication address and determine the target video frame from the video stream. Wherein, the detail extraction module may adopt a reading method of direct reading, or may adopt a reading method of indirect reading through other modules, which is not limited in the embodiment of the present specification. In addition, for the specific implementation process of the detail extraction module, reference may be made to the content description of the relevant parts of the image processing method and the video processing method, which will not be repeated here.

Then, the detail generation module may use the first detail image as a basis, the selected local area may represent the local native detail in the first detail image, and statistical feature information can be obtained through statistical characteristics between pixels in the local area, and then Update the color information of the target pixel to obtain a second detailed image containing subtle image features. For the specific implementation process of the detail generation module, reference may be made to the content descriptions of the relevant parts of the image processing method and the video processing method, which will not be repeated here.

In the specific implementation, since more attention is paid to the high-frequency band pixel information when acquiring statistical feature information, the pixels in the second detailed image may be in a higher frequency band, which is prone to generate noise. In order to reduce the noise of the second detailed image, as shown in Fig. As shown in 12, the video processing system 120 may further include: a detail processing module 124.

The detail processing module 124 is located between the detail generation module 122 and the video synthesis module 123, and is adapted to perform noise reduction processing on the second detail image obtained by the detail generation module 122, and denoise the second detail image after the noise reduction processing. The second detail image is sent to the video synthesis module 123 . The video synthesis module 123 combines the second detail image after noise reduction processing and the target video frame to obtain a composite video frame.

In a specific implementation, the video processing system provided in this specification can process the image-enhanced video, and has better compatibility and adaptability.

For example, as shown in FIG. 12 , the video processing system 120 may further include: a resolution processing module 125 . The resolution processing module 125 is respectively connected with the detail extraction module 121 and the video synthesis module 123, and is adapted to perform resolution amplification processing on the target video frame, and send the target video frame after resolution amplification processing. to the detail extraction module 121 and the video synthesis module 123 . The detail extraction module 121 extracts the image features of the target video frame after resolution enlargement processing to obtain a first detail image; the image synthesis module 123 combines the second detail image and the resolution enlargement processing target video frame, Get the composite video frame.

In a specific implementation, the resolution processing module can obtain the target video frame by reading the video frame data recorded on the designated storage address, or can read the video stream through the designated communication address and determine the target video frame from the video stream. The resolution processing module may adopt a reading method of direct reading, or may adopt a reading method of indirect reading through other modules, which is not limited in the embodiment of this specification.

In actual implementation, for the workflow of the video processing system, reference may be made to the related description of the workflow of the image processing system and FIG. 10 , which will not be repeated here.

The embodiment of this specification also provides a data processing module, the data processing module is applied to the laser radar, and is connected to the receiving part of the laser radar,

The data processing module may include a memory and a processor, the memory is suitable for storing one or more computer instructions, and the processor executes the image processing method or the video processing method described in any one of the foregoing embodiments when the processor executes the computer instructions. A step of. For specific steps, reference may be made to the foregoing embodiments, which will not be repeated here.

Optionally, the processor can be implemented by processing chips such as CPU (Central Processing Unit, central processing unit), GPU (Graphics Processing Unit, graphics processor), FPGA (Field Programmable Gate Array, field programmable logic gate array). , can also be implemented by an ASIC (Application Specific Integrated Circuit, a specific integrated circuit) or one or more integrated circuits configured to implement the embodiments of the present specification.

Optionally, the memory may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.

In a specific implementation, the data processing device may further include an extension interface, which is suitable for connecting with other devices to realize data interaction. The data processing device as described may be connected to a display device to play a composite image or a video stream containing composite video frames.

An embodiment of the present invention further provides a computer-readable storage medium, which stores computer instructions, which can execute the steps of the image processing method or the video processing method described in any of the foregoing embodiments of the present invention when the computer instructions are executed. Wherein, the computer-readable storage medium may be various suitable readable storage mediums such as an optical disc, a mechanical hard disk, and a solid-state hard disk. The instructions stored on the computer-readable storage medium execute the steps of the image processing method or the video processing method described in any of the foregoing embodiments, and the specific reference may be made to the foregoing embodiments, which will not be repeated.

The computer-readable storage medium may include, for example, any suitable type of memory unit, storage device, storage item, storage medium, storage device, storage item, storage medium and/or storage unit, eg, memory, removable or non- Removable media, erasable or non-removable media, writable or rewritable media, digital or analog media, hard disks, floppy disks, compact disc read only memory (CD-ROM), compact disc recordable (CD-R), Compact Disc Rewritable (CD-RW), Optical Disc, Magnetic Media, Magneto-Optical Media, Removable Memory Card or Disk, Various Types of Digital Versatile Disc (DVD), Magnetic Tape, Cassette, etc.

Computer instructions may include any suitable type of code implemented using any suitable high-level, low-level, object-oriented, visual, compiled and/or interpreted programming language, eg, source code, compiled code, interpreted code, executable code Execute code, static code, dynamic code, encrypted code, etc.

It should be noted that the terms "first" and "second" in the embodiments of the present specification are only used for description purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may expressly or implicitly include one or more of that feature. Also, the terms "first," "second," etc. are used to distinguish between similar objects and are not necessarily used to describe a particular order or precedence. It is to be understood that the terms so used are interchangeable under appropriate circumstances to enable the embodiments of the specification described herein to be practiced in sequences other than those illustrated or described herein.

Although the embodiments of the present disclosure are disclosed above, the embodiments of the present disclosure are not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the embodiments of the present disclosure. Therefore, the protection scope of the embodiments of the present disclosure should be based on the scope defined by the claims.

Claims

An image processing method, comprising:

extracting image features of the original image to obtain a first detail image;

Select target pixels and a local area from the first detail image, and the local area includes the target pixels;

Based on the statistical relationship between other pixels in the local area and the target pixel, the corresponding statistical feature information is calculated and obtained, and the statistical feature information is updated to the color information of the target pixel in the first detail image to obtain second detail image;

The second detail image and the original image are merged to obtain a composite image.
The image processing method according to claim 1, wherein before the combining the second detail image and the original image, the method further comprises:

A noise reduction process is performed on the second detail image, so as to combine the noise reduction processed second detail image and the original image.
The image processing method according to claim 2, wherein the performing noise reduction processing on the second detail image comprises:

performing guided filtering processing on the second detail image based on the original image;

Based on the preset modulated image, image modulation processing is performed on the second detail image, and the modulated second detail image and the original image are combined to obtain a composite image.
The image processing method according to claim 1, wherein the selecting the target pixel and the local area from the first detail image comprises:

Based on a preset selection condition, the target pixel is selected from the first detail image, and a local area including the target pixel is selected based on a preset area geometric parameter.
The image processing method according to claim 1, wherein, calculating and obtaining corresponding statistical feature information based on the statistical relationship between other pixels in the local area and the target pixel, comprising:

Acquire the color information of other pixels in the local area, and based on preset color conditions, determine the pixels that meet the color conditions from the local area, and obtain the Statistical feature information corresponding to the target pixel;

Wherein, the color condition is set according to the original color information of the target pixel.
The image processing method according to claim 1, wherein before the extracting the image features of the original image, the method further comprises:

Perform resolution enlargement processing on the original image to extract image features of the original image after resolution enlargement processing, and combine the second detail image and the original image after resolution enlargement processing.
The image processing method according to claim 1, wherein the extracting the image features of the original image comprises:

Based on the preset filter window geometric parameters and filter coefficients, the original image is filtered to extract the image features of the original image.
A video processing method, comprising:

Extracting the image features of the target video frame in the video stream to obtain the first detail image;

Select target pixels and a local area from the first detail image, and the local area includes the target pixels;

Based on the statistical relationship between other pixels in the local area and the target pixel, the corresponding statistical feature information is calculated and obtained, and the statistical feature information is updated to the color information of the target pixel in the first detail image to obtain second detail image;

The second detailed image and the target video frame are combined to obtain a composite video frame.
video processing method according to claim 8, is characterized in that, before described merging described second detail image and described target video frame, also comprises:

Perform noise reduction processing on the second detail image, so as to combine the second detail image after noise reduction processing with the target video frame.
The video processing method according to claim 9, wherein the performing noise reduction processing on the second detail image comprises at least one of the following:

performing guided filtering processing on the second detail image based on the target video frame;

Based on the preset modulated image, image modulation processing is performed on the second detailed image, and the modulated second detailed image and the target video frame are combined to obtain a composite video frame.
The video processing method according to claim 8, wherein the selecting a target pixel and a local area from the first detail image comprises:

Based on a preset selection condition, the target pixel is selected from the first detail image, and a local area including the target pixel is generated based on a preset geometric parameter of the area.
The video processing method according to claim 8, wherein, calculating and obtaining corresponding statistical feature information based on the statistical relationship between other pixels in the local area and the target pixel, comprising:

Acquire the color information of other pixels in the local area, and based on preset color conditions, determine the pixels that meet the color conditions from the local area, and obtain the Statistical feature information corresponding to the target pixel;

Wherein, the color condition is set according to the original color information of the target pixel.
The video processing method according to claim 8, wherein before extracting the image feature of the target video frame in the video stream, the method further comprises:

A resolution amplification process is performed on the target video frame to extract image features of the target video frame after the resolution amplification process, and the second detail image and the resolution amplification process target video frame are combined.
The video processing method according to claim 8, wherein the extracting the image features of the target video frame in the video stream comprises:

Based on the preset filter window geometric parameters and filter coefficients, filtering is performed on the target video frame to extract image features of the target video frame.
An image processing system, comprising:

a detail extraction module, suitable for extracting image features of the original image to obtain a first detail image;

A detail generation module, adapted to select a target pixel and a local area from the first detail image, the local area includes the target pixel, and based on statistics between other pixels in the local area and the target pixel relationship, calculate to obtain the corresponding statistical feature information, update the statistical feature information to the color information of the target pixel in the first detailed image, and obtain a second detailed image;

The image synthesis module is adapted to combine the second detailed image and the original image to obtain a composite image.
The image processing system according to claim 15, further comprising:

A detail processing module, located between the detail generation module and the image synthesis module, is adapted to perform noise reduction processing on the second detail image obtained by the detail generation module, and send the second detail image after noise reduction processing to the image synthesis module.
The image processing system according to claim 15, further comprising:

A resolution processing module, connected to the detail extraction module and the image synthesis module respectively, is adapted to perform resolution amplification processing on the original image, and send the original image after resolution amplification processing to the detail extraction module and the image synthesis module.
A video processing system, comprising:

a detail extraction module, adapted to extract the image features of the target video frame in the video stream to obtain the first detail image;

A detail generation module, adapted to select a target pixel and a local area from the first detail image, the local area includes the target pixel, and based on statistics between other pixels in the local area and the target pixel relationship, calculate to obtain the corresponding statistical feature information, update the statistical feature information to the color information of the target pixel in the first detailed image, and obtain a second detailed image;

The video synthesis module is adapted to combine the second detailed image and the target video frame to obtain a composite video frame.
The video processing system of claim 18, further comprising:

A detail processing module, located between the detail generation module and the video synthesis module, is adapted to perform noise reduction processing on the second detail image obtained by the detail generation module, and send the second detail image after noise reduction processing to the video synthesis module.
The video processing system of claim 18, further comprising:

A resolution processing module, connected to the detail extraction module and the video synthesis module respectively, is adapted to perform resolution amplification processing on the target video frame, and send the target video frame after resolution amplification processing to the details an extraction module and the video synthesis module.
A data processing device, comprising a memory and a processor, wherein the memory stores computer instructions that can run on the processor, wherein the processor executes claims 1 to 7 when running the computer instructions The steps of any one of or the methods of any one of claims 8 to 14.
A computer-readable storage medium on which computer instructions are stored, characterized in that, when the computer instructions are executed, the steps of the method described in any one of claims 1 to 7 or any one of claims 8 to 14 are executed.