Video enhancement method applied to wireless image transmission data chain system

Abstract

The invention discloses a video enhancement method applied to a wireless image transmission data chain system, which comprises the following steps: S1, a flight unit demodulates and decodes received data to obtain video data, converts the video data into an ordered sequence, and extracts a key frame of the video data by calculating a shot boundary coefficient; s2, performing enhancement processing on the key frame image, performing color compensation on a picture in the key frame image by using a spherical particle model, and enhancing the image quality of the key frame image; s3, the contrast of the key frame image is improved through a dynamic fuzzy histogram equalization image enhancement algorithm based on brightness maintenance, so that the key frame image is bright and dark, and target features are highlighted; and S4, performing enhancement processing on the key frame image by using a time domain fractional order differential algorithm, and enhancing the texture of the key frame image, so that scene details in the image are obvious and convenient to identify, and obtaining the enhanced key frame image. According to the invention, by performing enhancement processing on the airborne received video data, the enhanced video data is clearer in details and more obvious in features.

Description

Technical Field

[0001] The present invention relates to the field of video image processing, and in particular to a video enhancement method applied to a wireless image transmission data link system. Background Art

[0002] Airborne communication is gradually becoming an important supplement and reinforcement means of information system with its advantages of high data transmission efficiency and reliability. As an important part of airborne communication, wireless image transmission data link system has been widely used in the current wireless communication field, and the video enhancement technology in it will become the focus of future research. In order to ensure the smooth playback of real-time video transmission, the current wireless image transmission data link system performs corresponding anti-interference processing at the wireless link layer. This method can only ensure that the signal will not be interfered with during the wireless transmission process, but ignores the noise interference caused by the video shooting itself. In the actual battlefield environment, due to the limitations of shooting conditions, such as clouds, fog, dust and other bad weather conditions, the images captured by the sensor will have different degrees of distortion, noise and unclear details. At this time, when the video is transmitted back to the airborne terminal through the wireless link for playback, the blurred and unclear video images will cause the operator to be unable to accurately identify the content in the video, which will affect the operator's judgment of the target. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a video enhancement method applied to a wireless image transmission data link system. By enhancing the video data received on board, the details of the enhanced video data are clearer and the features are more obvious, which facilitates the operator to make a more accurate and clear judgment on the target.

[0004] The object of the present invention is to achieve the following technical solution: a video enhancement method applied to a wireless image transmission data link system, comprising the following steps: S1. The flight unit encodes and modulates the captured video data and sends it to the airborne terminal through the RF antenna. The airborne terminal demodulates and decodes the received data to obtain video data, converts the video data into an ordered sequence, and extracts the key frames of the video data by calculating the lens boundary coefficients; S2. Enhance the key frame image, use the spherical particle model to perform color compensation on the key frame image, and enhance the key frame image quality; S3. Improve the contrast of key frame images through dynamic fuzzy histogram equalization image enhancement algorithm based on brightness preservation, so that the key frame images are clear in light and dark and highlight the target features; S4. Use the time-domain fractional differential algorithm to enhance the key frame image, strengthen the texture of the key frame image, make the scene details in the image obvious, and facilitate identification, thereby obtaining an enhanced key frame image.

[0005] The beneficial effect of the present invention is that the present invention enhances the real-time video data received on board, so that the details of the enhanced real-time video data are clearer and the features are more obvious, making it more in line with human visual characteristics than the original video, and facilitating the operator to make more accurate and clear judgments on the target. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 is a flow chart of the method of the present invention; Figure 2 Schematic diagram of the video organization structure. DETAILED DESCRIPTION

[0007] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0008] like Figure 1 As shown, a video enhancement method applied to a wireless image transmission data link system comprises the following steps: S1. The flight unit encodes and modulates the captured video data and sends it to the airborne terminal through the RF antenna. The airborne terminal demodulates and decodes the received data to obtain video data, converts the video data into an ordered sequence, and extracts the key frames of the video data by calculating the lens boundary coefficients; The flight unit includes a camera and a flight data link. The airborne terminal is placed on the ground. The camera inputs the captured data into the flight data link. The flight data link sends the data to the airborne terminal wirelessly. The airborne terminal parses the received wireless data into video. In complex environments, due to the influence of clouds, fog, and dust, the atmospheric visibility is reduced, and the color of the objects captured by the sensor is attenuated to a certain extent, causing the captured scene to be blurred and difficult to identify. The operator cannot accurately obtain the target features in the video, so it is necessary to enhance the video transmitted back in complex environments.

[0009] The first step is to extract the video key frames. The video data is organized according to its structure (organization structure such as Figure 2As shown in the figure, it can be divided into an ordered sequence of scenes, shots and frames. Since there is a lot of redundancy between the frames of the same shot, it is only necessary to select one or more frames as key frames to represent the shot. By extracting the key frames, the complete content of the video can be effectively reflected. The purpose of doing so is to minimize the redundancy of video data and facilitate subsequent video enhancement processing. The present invention adopts a key frame extraction technology based on shot boundaries. The principle is to calculate the shot boundary coefficient through a shot boundary detection algorithm, and then adaptively extract the key frames in the shot according to the distribution of the shot boundary coefficient. This method is easy to implement and has low complexity, and has little effect on delay. The number of frames for adaptively extracting key frames can also change autonomously according to the scene. When there are many shot scenes, the sampling frequency of extracting key frames is increased, and when the shot scene is single, the sampling frequency of extracting key frames is reduced.

[0010] It should be noted that video data is an ordered sequence with a large amount of redundant information. According to its organizational structure, it can be subdivided into scenes, shots and frames. The neighborhood window frame is all frames within a width of 2N+1 (assuming that the video data is a frame sequence of 1, 2, 3, 4, 5, 6, 7... and the neighborhood window width is 3, then the neighborhood window frame is all frames within 3, that is, 1, 2, 3) Assuming the width of the neighborhood window is 2N+1, the calculation formula for the neighborhood window frame difference of the i-th frame is (1) The calculation formula of the shot boundary similarity coefficient of the i-th frame is: (2) The lens boundary coefficient of the i-th frame (3) In the formula, , is the frame difference between the mth frame and the nth frame; is the shot boundary detection operator, and its coefficient is proportional to the neighborhood window frame difference. a is a constant, ranging from 5 to 15.

[0011] After calculating the lens boundary coefficient, the boundary range of the current lens is obtained. It can be considered that the data frames within the boundary range are all data frames of the current lens. At the same time, the data frames within the range are sampled, that is, key frame extraction. In the embodiment of the present application, the number of key frames extracted is related to the complexity of the lens content. Generally, the variance of the frames within the lens is calculated. The larger the variance, the more key frames are extracted. In the present application, the number of extractions can be regarded as a parameter preset based on experience.

[0012] S2. Enhance the key frame image, use the spherical particle model to perform color compensation on the key frame image, and enhance the key frame image quality; After determining the key frame, color compensation is performed on the key frame image. In complex weather conditions, dust or water droplets floating in the atmosphere will change the refractive index, causing the reflected light of the target object to interact with it to form scattering, which ultimately causes electromagnetic wave attenuation, thereby causing the color of the object photographed by the sensor to be distorted. Since visible light is composed of red, green, and blue, the red, green, and blue light of each pixel can be compensated to achieve the purpose of enhancing the video quality. Using the spherical example model, the attenuation coefficient expression can be calculated as: (4) in is the atmospheric visibility, is the wavelength; a is the wavelength correction factor, under different conditions a The value of is: (5) The correction values ​​for visibility below 6 km are as follows: (6) Due to the influence of scattering on electromagnetic waves, the energy of infrared radiation is weakened. The weakening of its energy is related to the atmospheric transmittance. Therefore, the attenuation coefficient of electromagnetic wave propagation per unit distance is: (7) The above formula can be used to obtain the corresponding attenuation equation through integration operation: (8) in is the radiation intensity value, and is the distance value, the exponential term is the corresponding refractive index, and the attenuated color component is calculated by the attenuation equation , and finally the enhanced color is obtained by the following formula: represents the image color after compensation, where Represents the three colors red, green and blue.

[0013] S3. Improve the contrast of key frame images through dynamic fuzzy histogram equalization image enhancement algorithm based on brightness preservation, so that the key frame images are clear in light and dark and highlight the target features; After color compensation is completed, a dynamic fuzzy histogram equalization image enhancement algorithm based on brightness preservation is used. This algorithm can effectively process noisy images and improve image contrast.

[0014] The histogram of the digital image processed in step S2 is converted into a fuzzy histogram. for The fuzzy pixel value is expressed as follows: (9) in is the triangular membership function, is the support domain of the membership function; Calculate the peak point of the fuzzy histogram and set the peak point as the threshold value to segment the histogram. Since the peak point is the maximum point of the fuzzy histogram, the peak point position can be obtained by calculating the first-order derivative and second-order derivative of the histogram curve. The first-order derivative is as follows: (10) Finding the second-order derivative is as follows: (11) Depend on , we can get the maximum point set of the histogram curve as: (12) According to the peak set p, the histogram can be divided into multiple sub-histogram areas. ,in Map each area of ​​the fuzzy histogram to a dynamic range: (13) in and They are the maximum and minimum intensities of the subhistograms, is the total number of pixels, Adjust the input subhistogram, Determines the dynamic range of the output subhistogram.

[0015] (14) (15) thus The grayscale area range of the sub-histogram is .

[0016] Finally, each sub-region is balanced, namely: (16) in, is the intensity of the kth fuzzy histogram, is the cumulative distribution function; S4. Use the time-domain fractional differential algorithm to enhance the key frame image, strengthen the texture of the key frame image, make the scene details in the image obvious, and facilitate identification, thereby obtaining an enhanced key frame image.

[0017] After the image contrast is improved, the time-domain fractional differential algorithm is used to enhance the image, making the detail features more prominent by improving the image texture clarity; Set signal , then the signal of The derivative can be: (17) The fractional-order partial differential equation of a two-dimensional digital image is: (18) (19) Take the eight symmetrical points of a certain pixel in the image as the mask operator. Let the coefficients of the mask operator be , then the center point coefficient is , the coefficients of the 8 symmetric point directions can be obtained from the above two formulas: (20) Using this mask operator Image enhancement is achieved by performing spatial convolution filtering on the digital image to be enhanced, and the calculated operator is used as the filter coefficient during filtering.

[0018] The above is a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not depart from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. A video enhancement method applied to a wireless image transmission data link system, characterized in that: The following steps are involved: S1. The flight unit encodes and modulates the captured video data and sends it to the airborne terminal through the RF antenna. The airborne terminal demodulates and decodes the received data to obtain video data, converts the video data into an ordered sequence, and extracts the key frames of the video data by calculating the lens boundary coefficients; S2. Enhance the key frame image, use the spherical particle model to perform color compensation on the key frame image, and enhance the key frame image quality; S3. Improve the contrast of key frame images through dynamic fuzzy histogram equalization image enhancement algorithm based on brightness preservation, so that the key frame images are clear in light and dark and highlight the target features; S4. Use the time-domain fractional differential algorithm to enhance the key frame image, strengthen the texture of the key frame image, make the scene details in the image obvious, and facilitate identification, thereby obtaining an enhanced key frame image.

2. The video enhancement method applied to a wireless image transmission data link system according to claim 1, characterized in that: The method further comprises: exporting the enhanced key frame image synthesized video data obtained in step S4 to a terminal for display.

3. The video enhancement method applied to a wireless image transmission data link system according to claim 1, characterized in that: In the step S1, extracting the key frame of the video data by calculating the shot boundary coefficient includes: Assume that the width of the neighborhood window is 2N+1, then the neighborhood window frame difference of the i-th frame is The calculation formula is (1) Shot boundary similarity coefficient of the i-th frame Calculation formula: (2) The lens boundary coefficient of the i-th frame (3) In the formula, , is the frame difference between the mth frame and the nth frame; is the shot boundary detection operator, whose coefficient is proportional to the neighborhood window frame difference. a is a constant; After calculating the lens boundary coefficient, the boundary range of the current lens is obtained. It is considered that all data frames within the boundary range are data frames of the current lens. At the same time, the data frames within the range are sampled, that is, key frame extraction.

4. The video enhancement method for a wireless image transmission data link system according to claim 1, characterized in that: In step S2, the purpose of enhancing the video quality is achieved by compensating the red, green and blue light of each pixel. Using the spherical example model, the attenuation coefficient expression is calculated as: (4) in is the atmospheric visibility, is the wavelength; a is the wavelength correction factor, under different conditions a The value of is: (5) The correction values ​​for visibility below 6 km are as follows: (6) Due to the influence of scattering on electromagnetic waves, the energy of infrared radiation is weakened. The weakening of its energy is related to the atmospheric transmittance. Therefore, the attenuation coefficient of electromagnetic wave propagation per unit distance is: (7) The above formula can be used to obtain the corresponding attenuation equation through integration operation: (8) in is the radiation intensity value, and is the distance value, the exponential term is the corresponding refractive index, and the attenuated color component is calculated by the attenuation equation , and finally the enhanced color is obtained by the following formula: represents the image color after compensation, where Represents the three colors red, green and blue.

5. The video enhancement method applied to a wireless image transmission data link system according to claim 1, characterized in that: The step S3 comprises: The histogram of the digital image processed in step S2 is converted into a fuzzy histogram. for The fuzzy pixel value is expressed as follows: (9) in is the triangular membership function, is the support domain of the membership function; Calculate the peak point of the fuzzy histogram, set the peak point as the threshold value to segment the histogram; since the peak point is the maximum point of the fuzzy histogram, the peak point position is obtained by taking the first-order derivative and the second-order derivative of the histogram curve. The first-order derivative is as follows: (10) Finding the second-order derivative is as follows: (11) Depend on , the maximum point set of the histogram curve is obtained as: (12) According to the peak set p, the histogram is divided into multiple sub-histogram areas. ,in Map each area of ​​the fuzzy histogram to a dynamic range: (13) in and They are the maximum and minimum intensities of the subhistograms, is the total number of pixels, Adjust the input subhistogram, Determines the dynamic range output to the sub-histogram; (14) (15) thus The grayscale area range of the sub-histogram is ; Finally, each sub-region is balanced, namely: (16) in, is the intensity of the kth fuzzy histogram, is the cumulative distribution function, Indicates the result of equalization of the subhistogram.

6. The video enhancement method for a wireless image transmission data link system according to claim 1, characterized in that: The step S4 comprises: Set signal , then the signal of The derivative is: (17) The fractional-order partial differential equation of a two-dimensional digital image is: (18) (19) Take the eight symmetrical points of a certain pixel in the image, namely, up, down, left, right, upper left, lower left, upper right, and lower right as the mask operator; let the coefficient of the mask operator be , then the center point coefficient is , there are 8 coefficients in the direction of symmetric points, let n=8, then: (20) Using this mask operator Image enhancement is achieved by performing spatial convolution filtering on the digital image to be enhanced, and the calculated operator is used as the filter coefficient during filtering.

Images