A front-looking sonar image pseudo-color processing method for improving turbidity influence of water body

Abstract

This invention discloses a pseudo-color processing method for forward-looking sonar images to mitigate the impact of water turbidity. The core of this method lies in dynamically determining a threshold based on the statistical analysis of current sonar echo intensity and then linearly normalizing the statistical intensity range. A custom Jet color bar is generated based on the characteristics of the forward-looking sonar image, and this custom Jet color bar is dynamically adjusted according to different water turbidity levels to achieve pixel-level adaptive color mapping. This method can significantly increase the visualization quality and target recognition of sonar images in water environments with varying turbidity, improving the operator's perception and judgment of underwater targets. It overcomes the detail loss problem caused by existing technologies that only use fixed pseudo-color mapping, has broad application prospects, and requires only software implementation, possessing good compatibility and promotional value.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology and relates to an image pseudo-color processing method. It dynamically sets the color mapping threshold by statistical echo intensity and adaptively adjusts the Jet color bar by combining water turbidity data to achieve pseudo-color image enhancement display. Background Technology

[0002] The development of underwater detection and operation technologies has become a key area of ​​focus both domestically and internationally. Forward-looking sonar, with its core advantage of outputting high-resolution echo images with a fan-shaped field of view underwater, is widely used in underwater target detection scenarios. Forward-looking sonar equipment detects the ocean through sound waves, generating data sequences by collecting multiple sound wave intensities. However, raw sonar images often suffer from low resolution, severe speckle noise, and unclear features. When presented directly as grayscale images, the information expression is limited, and the human eye's ability to distinguish grayscale levels is poor, making it difficult to effectively identify details such as underwater targets, obstacles, or topography. When forward-looking sonar detects in environments with high water turbidity, the scattering and absorption of a large number of suspended particles result in many weak echo data, leading to a significant increase in background noise intensity. The rendered image will then display more dense "snowflakes" or noise.

[0003] To enhance the visualization and information recognition of sonar images, pseudo-color processing techniques are widely used. This technique maps the monotonous grayscale value of sonar echo intensity to a color space more sensitive to the human eye, using different colors to distinguish different echo intensities, thereby highlighting image details and depth. Among these, Jet color mapping has become a commonly used pseudo-color mapping scheme due to its high contrast and natural color transitions. The literature "AUV Target Recognition and Tracking Based on Forward-Looking Sonar Images" provides a detailed description of the standard Jet color mapping implementation, demonstrating that Jet color mapping has been widely used for pseudo-colorization of forward-looking sonar images. Experimental results prove that Jet color mapping can improve visual contrast and target recognition. The patent "An Image Quality Optimization Method for Forward-Looking Sonar" proposes a quality optimization process for forward-looking sonar images, involving the implementation details of image preprocessing, dynamic compression, and color mapping, and points out the advantages of Jet in contrast and detail rendering. Existing literature provides detailed descriptions of the specific implementation details of Jet color mapping, verifying the maturity of forward-looking sonar Jet color mapping technology. However, current literature only demonstrates the advantages of standard Jet color mapping in pseudo-color processing of forward-looking sonar images, and has not optimized Jet color mapping for the characteristics of forward-looking sonar detection images and water turbidity.

[0004] In summary, existing pseudo-color processing techniques for forward-looking sonar images lack the ability to adapt to the inherent statistical characteristics of sonar echo signals and changes in the external aquatic environment, which limits their effectiveness in complex and ever-changing marine environments. There is an urgent need for an intelligent processing method that can dynamically optimize color mapping strategies. Summary of the Invention

[0005] In view of this, the present invention provides a pseudo-color processing method for forward-looking sonar images to improve the effect of water turbidity. The method dynamically adjusts the pseudo-color mapping strategy based on the real-time echo intensity distribution of the forward-looking sonar and the water turbidity to generate a custom grayscale-color mapping, thereby significantly improving the clarity, contrast and information recognizability of sonar images in different scenes and environments.

[0006] The specific technical solution is as follows: A method for processing pseudo-color in forward-looking sonar images to mitigate the impact of water turbidity, specifically including: Based on the standard Jet color bar, a custom Jet color bar suitable for forward-looking sonar images is generated. Then, the custom Jet color bar is dynamically adjusted according to the turbidity of different water environments to generate the optimal Jet color bar for the current water turbidity environment. The sonar image is converted into a pseudo-color image using the optimized Jet color bar mapping relationship.

[0007] Furthermore, the process of generating the optimized Jet color bar includes generating a custom Jet color bar based on the standard Jet color bar, treating the custom color bar as a set of color arrays, the number of which is... The array elements are The value, x, is the array index, and its range is [value]. ), Quantize the red, green, and blue channels corresponding to x; obtain the water turbidity value T normalized to the [0,1] interval, where 0 represents normal water and 1 represents extremely turbid water; adjust the color distribution of the Jet color bar through parameter T to generate the Jet color bar optimized for the current water turbidity environment. The optimization color bar adjustment logic is as follows: the echo data intensity of the forward-looking sonar detection target (e.g., mine) is located in the middle area of ​​the overall intensity range. When the water turbidity value is close to 0, the adjusted color bar distribution is consistent with the custom Jet color bar; when the value of T is within (0,1], the color of the middle area is still green and yellow, and the proportion of this color system in the color bar is larger than that of the standard Jet color bar. This ensures that the target clarity and brightness are basically consistent with the effect of using the custom Jet color mapping, while increasing the proportion of black and blue colors in the color bar, so that more noise data is mapped to black or blue, and the brightness of noise colors is reduced. Furthermore, the specific implementation of the Jet color mapping is as follows: the number of color arrays is set to... The Jet color mapping is provided by A one-dimensional array consisting of color nodes Let x be the color array index, with a value range of... Iterate through x to generate a set of custom color arrays When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When implementing the above color array in a computer language, Each channel value will be limited to [0,1]. Channel values ​​greater than 1 will be set to 1 by default, and channel values ​​less than 0 will be set to 0 by default.

[0008] Furthermore, the sonar image to pseudo-color image conversion specifically involves: acquiring sonar images, statistically analyzing the forward-looking sonar echo intensity in real time, and dynamically setting a color mapping threshold based on the statistical results; using the color mapping threshold as a dynamic range, normalizing the grayscale value of each pixel G(i,j) in the sonar image to... The normalized pixel values ​​are obtained from the interval. ; Normalized pixel values Assign the value to the color index x, and pass it in. Find the corresponding RGB color value; assign the obtained RGB color value to the pixel at the corresponding position in the output pseudocolor image.

[0009] Furthermore, the maximum value and minimum value A color mapping threshold is set, which is updated in real time based on the acquired grayscale image to adapt to changes in the environment.

[0010] Beneficial effects 1. Enhanced image contrast and detail: By dynamically setting thresholds based on real-time echo intensity statistics, it can adaptively and optimally stretch the echo intensity of the current scene, avoiding images that are too dark or too bright due to scene changes, increasing color contrast, and greatly improving the effective information content of the image.

[0011] 2. This method incorporates water turbidity as a key parameter into the sonar pseudo-color processing workflow. Through the parametric design of a custom Jet color bar, the original gradient colors are preserved while the color distribution is flexibly adjusted according to different water environments or mission requirements, thereby improving the visualization effect of the image.

[0012] 3. The color mapping process has low computational cost, making it suitable for real-time processing and display of forward-looking sonar images.

[0013] 4. This method is compatible with existing forward-looking sonar hardware, is implemented only at the software level, and has good engineering portability.

[0014] 5. The color mapping function uses piecewise interpolation to ensure smooth color transitions. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the method.

[0016] Figure 2 This is a sonar image processed using the standard Jet color bar pseudo-color method.

[0017] Figure 3 This method provides a custom Jet color bar pseudo-color processing sonar image.

[0018] Figure 4 Turbidity The sonar image after pseudo-color processing using this method. Detailed Implementation

[0019] This invention provides real-time statistical analysis of forward-looking sonar echo intensity and dynamically sets color mapping thresholds based on the statistical results. It generates a custom Jet color bar suitable for forward-looking sonar images based on the standard Jet color bar, and then dynamically adjusts the custom Jet color bar according to the turbidity of different water environments. Normalized intensity values ​​are mapped to the generated custom Jet color bar to generate pseudo-color sonar images. Threshold parameters are updated in real-time based on the acquired grayscale images to adapt to environmental changes. The preset color mapping scheme is the custom Jet color bar. The custom Jet color bar includes six color gradients, and the proportion of each color is controlled by the user. The color bar is represented as a set of color nodes on the intensity values. A non-linear adjustment function is established, taking water turbidity data and the custom Jet color bar as inputs. Based on the current water turbidity data, the colors on the intensity axis are changed to generate the custom color bar. Under high turbidity conditions, the color bar ensures that the color system of the target to be distinguished remains unchanged while increasing the proportion of black and blue colors in the color bar, so that more noise data is mapped to black or blue, reducing the brightness of noise colors. The color mapping function is implemented using piecewise interpolation to ensure smooth color transitions.

[0020] The following will provide a detailed description: The specific steps of this method are as follows: 1. Generate a color array for a custom Jet color bar. The number of colors in the array is (Generally set to 343), it consists of a series of Value composition, where x is the color array index, and the value range is... Divide the range of values ​​for index x into 6 sub-ranges, and iterate through x to generate a set of custom one-dimensional arrays. When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as .

[0021] The standard Jet color bar is divided into five segments: [0, 1 / 8] is blue, [1 / 8, 3 / 8] is a gradient from blue to cyan, [3 / 8, 5 / 8] is a gradient from cyan to yellow, [5 / 8, 7 / 8] is a gradient from yellow to red, and [7 / 8, 1] is red. Based on the analysis of a large number of forward-looking sonar raw grayscale images, strong echo data usually manifests as interference (interference caused by the hull, other acoustic equipment, etc.) and terrain. The echo data intensity of the target or obstacle to be distinguished is located in the middle area of ​​the overall intensity range. The middle area of ​​the standard Jet color bar is cyan and accounts for a very small proportion. After pseudo-color mapping, the brightness and clarity of the target are not high.

[0022] Therefore, the custom color bar re-divided the intervals, changing the five segments of the standard Jet color bar to six segments: [0, 1 / 8] is a black to blue gradient, [1 / 8, 2 / 8] is a blue to cyan gradient, [2 / 8, 3 / 8] is a cyan to green gradient, [3 / 8, 5 / 8] is a green to yellow gradient, [5 / 8, 7 / 8] is a yellow to red gradient, and [7 / 8, 1] is red. In the [0, 1 / 8] interval, the custom color bar increases the proportion of black by changing the degree of color gradient. The color contrast of the custom color bar after pseudo-color rendering is higher than that of the standard color bar. In the [1 / 8, 5 / 8] interval, the proportion of green is increased and the proportion of blue is decreased. In the [3 / 8, 5 / 8] interval, the proportion of yellow is increased and the proportion of red is decreased. This makes the central area of ​​the custom color bar green and yellow, increasing the clarity and brightness of the target pseudo-color rendering.

[0023] Of course, other custom Jet color bars can also be generated as needed; the above is just one implementation method.

[0024] 2. Obtain the turbidity value of the water body in the current detection environment. The system operator manually sets a turbidity level (e.g., low, medium, high) based on experience or external information, and maps it to a numerical parameter. .

[0025] 3. Based on the current turbidity value of the water body (already normalized to) (A range, where 0 represents normal water and 1 represents extremely turbid water), generating a... Color array: Set the number of colors in the array. The Jet color mapping is provided by A one-dimensional array consisting of color nodes Let x be the color array index, with a value range of... Iterate through x to generate a set of custom color arrays When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as .

[0026] Based on the analysis of a large number of raw grayscale images from forward-looking sonar, in environments with high water turbidity, the scattering and absorption of numerous suspended particles result in a significant increase in weak echo data, leading to a substantial increase in background noise. The rendered image will then display more dense "snowflakes" or noise. The echo data intensity of forward-looking sonar targets (such as mines) lies in the middle region of the overall intensity range. When the water is clear... , The color bar is close to the custom Jet color bar distribution. When the value of T is within (0, 1), the color of the middle area remains green and yellow, and the proportion of this color system in the color bar is larger than that of the standard Jet color bar. This ensures that the target clarity and brightness are basically consistent with the effect of using custom Jet color mapping, while increasing the proportion of black and blue color systems in the color bar, so that more noise data is mapped to black or blue, and the brightness of noise colors is reduced. 4. Real-time color mapping threshold for sonar echo intensity: Acquire a frame of sonar image data, traverse all pixels of the frame, and calculate the maximum grayscale value. and minimum value ,Will and Set as the color mapping threshold; 5. Finally, convert the grayscale image... Convert to pseudo-color image For each pixel G(i,j) in G, perform the following mapping logic: Normalization: First, obtain the number of colors in the Jet color bar. , will pixels by The dynamic threshold range is normalized to Within the range, its normalization function is:

[0027] The normalized grayscale image has the following pixel count: ; Color search: The intensity value Use the index to find its corresponding RGB color value.

[0028] Pixel assignment: Assign RGB values ​​to the output pseudo-color image. corresponding pixels .

[0029] After traversing all pixels, an adaptively enhanced pseudo-color forward-looking sonar image is generated, which is then presented to the user via a display terminal.

[0030] Figure 3 This is a sonar image processed with a custom Jet color bar pseudo-color, for comparison. Figure 2 As can be seen, the image has higher color contrast, and the weak echo data is darker after rendering, making the target stand out better against the darker background. After rendering with the custom color bar, the target outline is clearer, the brightness is stronger, and it is easier to distinguish with the naked eye. Figure 4 This is a sonar image after pseudo-color processing with added turbidity, compared to... Figure 3 It can be seen that when turbidity At that time, the image after Jet color bar mapping, the target shown in the box, has basically the same clarity and brightness as the effect of using custom Jet color mapping, but the color of the interference noise in the image is darker, the overall image clarity is higher, and the visual effect is better.

[0031] Although embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make several modifications without departing from the principles of the present invention, and these modifications should also be considered to fall within the scope of protection of the present invention.

Claims

1. A method for processing pseudo-color in forward-looking sonar images to improve the effect of water turbidity, characterized in that: Based on the standard Jet color bar, a custom Jet color bar suitable for the target echo characteristics of forward-looking sonar images is generated. Then, the custom Jet color bar is dynamically adjusted according to the turbidity of different water environments to generate the optimal Jet color bar for the current water turbidity environment. The sonar image is converted into a pseudo-color image using the optimized Jet color bar mapping relationship.

2. The method according to claim 1, characterized in that: The process of generating the optimized Jet color bar includes constructing a custom Jet color bar based on the standard Jet color bar, treating the custom color bar as a set of color arrays, with the number of arrays being... The array elements are The value, x, is the array index, and its range is [value]. ), Quantize the red, green, and blue channels corresponding to x; acquire water turbidity data and normalize it to the [0,1] interval to obtain the water turbidity value T, where T=0 represents normal water and T=1 represents extremely turbid water; based on the analysis results of a large number of forward-looking sonar raw grayscale images, dynamically adjust the color distribution of the custom Jet color bar through the water turbidity value T to generate the optimal Jet color bar adapted to the current water environment. The adjustment logic is as follows: the echo data intensity of the detected target is always located in the middle area of ​​the overall intensity range. Regardless of the value of T, the mapping color corresponding to this middle area remains green and yellow, and the proportion of this color system in the color bar is greater than that of the standard Jet color bar, ensuring the clarity and brightness of the target are stable; when T=0, the color distribution of the optimal Jet color bar and the custom Jet color bar are... The color bars are completely consistent; when T∈(0,1], the proportion of black and blue colors in the color bars is increased, so that more weak echo noise data generated by the scattering or absorption of suspended particles in the water are mapped to the black or blue area, weakening the visual brightness of the noise and reducing interference with target recognition.

3. The method according to claim 2, characterized in that: The specific implementation of optimizing the Jet color bar mapping relationship is as follows: The number of colors in the array is set to... The optimized Jet color bar mapping relationship is determined by... A one-dimensional array consisting of color nodes Let x be the color array index, with a value range of... Iterate through x to generate a set of custom color arrays When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as When x iterates through the values ​​in Within the interval, The value is represented as .

4. The method according to claim 3, characterized in that: The sonar image to pseudo-color image conversion specifically involves acquiring sonar images, statistically analyzing the forward-looking sonar echo intensity in real time, and dynamically setting a color mapping threshold based on the statistical results. Using the aforementioned color mapping threshold as the dynamic range, the grayscale value of each pixel G(i,j) in the sonar image is normalized to... The interval is used to obtain the normalized pixel values. ; Normalized pixel values Assign the value to the color index x, and substitute it with... Find the corresponding RGB color value; assign the obtained RGB color value to the pixel at the corresponding position in the output pseudocolor image.

5. The method according to any one of claims 1-4, characterized in that: Take the maximum value of the image grayscale. and minimum value A color mapping threshold is set, which is updated in real time based on the acquired grayscale image to adapt to changes in the environment.

Images