A method for adaptive image enhancement of a television camera
By adaptively adjusting the grayscale range and Gamma correction parameters, the problems of insufficient scene adaptation and color imbalance in digital television imaging are solved, achieving high-quality image enhancement, improving image contrast and detail, and avoiding color distortion.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUNAN HUANAN OPTOELECTRONIC GRP CO LTD
- Filing Date
- 2026-03-11
- Publication Date
- 2026-06-12
AI Technical Summary
Existing image enhancement algorithms suffer from insufficient scene adaptation and color and brightness imbalance in digital television imaging. Traditional methods are prone to over-enhancing images, amplifying noise, and color distortion, while deep learning-based methods lack robustness and have high hardware requirements.
By adopting an adaptive adjustment method for grayscale range and Gamma correction parameters, and by calculating pixel histograms and grayscale mean values, differentiated nonlinear mapping rules are determined, a segmented pixel mapping model is constructed, and adaptive enhancement is performed in combination with the HSV color model to avoid image distortion.
It achieves high-quality image enhancement in different scenarios, improves image contrast and detail, avoids color distortion, adapts to images with different grayscale distributions, and improves the imaging quality of digital television.
Smart Images
Figure CN122199352A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of image processing technology, and specifically relates to an adaptive image enhancement method for television cameras. Background Technology
[0002] The image quality of digital television is directly related to the user's viewing experience. Color saturation, brightness, and contrast, as core indicators of image visual performance, directly affect the visual presentation of the image. However, due to limitations in the performance of devices such as cameras and mobile terminals, as well as environmental factors, insufficient light entering the acquisition equipment can lead to uneven brightness and darkness in the resulting image. This can result in low image contrast, loss of detail, and a decrease in image quality, significantly reducing the image's usability. Therefore, there is an urgent need for an adaptive enhancement technology that can adapt to all scenarios of digital television imaging and achieve coordinated optimization of color and brightness. This technology would address issues such as insufficient scene adaptation and color and brightness imbalance in existing technologies, thereby improving the image quality of digital television and the user's viewing experience.
[0003] Image enhancement is an important technique in image processing. Its purpose is to improve the visual effects of an image, such as contrast and detail clarity, through certain algorithms, so as to better meet the needs of subsequent image analysis, recognition, or visual observation.
[0004] Researchers have conducted extensive algorithmic research in the field of digital image processing, proposing various image enhancement algorithms from different perspectives. These algorithms not only improve the subjective visual effect of images but also enhance their overall quality. However, existing image enhancement algorithms still have some limitations:
[0005] Traditional image enhancement algorithms, such as linear enhancement (e.g., histogram equalization) and non-linear enhancement (e.g., gamma correction), have achieved considerable success in various fields after years of development. Histogram equalization improves contrast by redistributing pixel gray levels, but it can easily lead to over-enhancement, noise amplification, and color distortion. Traditional gamma correction requires manual setting of the gamma parameter, lacks adaptability, and is difficult to adapt to images with different gray-level distribution characteristics. When the gray-level distribution of an image varies greatly, fixed-parameter gamma correction cannot achieve the desired enhancement effect.
[0006] Deep learning-based image enhancement algorithms, particularly those based on supervised learning, offer excellent enhancement results but are highly dependent on the input dataset. Their robustness decreases when the scene changes. Unsupervised learning-based algorithms, while not requiring large input datasets, are significantly less effective than supervised learning algorithms. Furthermore, deep learning-based image enhancement algorithms have high hardware requirements, limiting their practical applications. Summary of the Invention
[0007] To address the aforementioned problems, the purpose of this invention is to provide an adaptive image enhancement method for television cameras, which can adaptively adjust the effective grayscale range and Gamma correction parameters according to the grayscale distribution characteristics of the image, thereby achieving high-quality image enhancement in different scenarios and improving the digital television imaging effect.
[0008] To achieve the above objectives, the present invention adopts the following technical solution: an adaptive image enhancement method for a television camera, comprising the following steps:
[0009] Step 1: Read the color image and separate the color image to obtain the pixel matrix of the three channels: red (R), green (G), and blue (B).
[0010] Step 2: Calculate the pixel histograms for the three channels: red (R), green (G), and blue (B). The pixel histograms are used to represent the number of pixels corresponding to each gray level (0-255).
[0011] Step 3: The average pixel value of each channel can be obtained by weighting the gray level and the corresponding pixel frequency. The numerical distribution can objectively reflect the brightness characteristics of the channel.
[0012] Step 4: For channels with different brightness characteristics, a differentiated nonlinear mapping rule is used to determine the Gamma parameter. At the same time, in order to avoid image distortion caused by extreme parameters, the range of Gamma parameter values is reasonably constrained to ensure that it is within the preset effective range.
[0013] Step 5: Calculate the average value of the Gamma parameter for the three channels as a unified Gamma correction parameter to ensure color consistency across the three channels and avoid color distortion.
[0014] Step 6: Count the total number of pixels in the image, and traverse the histogram to obtain the minimum and maximum non-zero gray values of each channel. These two values correspond to the actual pixel gray value distribution range of the original image.
[0015] Step 7: Forward cumulative histogram pixel frequency. When the cumulative number of pixels reaches the minimum threshold of the total number of pixels, the corresponding gray value is the low gray-scale cropping threshold. Reverse cumulative histogram pixel frequency. When the cumulative number of pixels exceeds the maximum threshold of the preset total number of pixels, the corresponding gray value is the high gray-scale cropping threshold. The two are the actual effective gray-scale range.
[0016] Step 8: Expand and correct the minimum / maximum non-zero grayscale value in Step 6, while preventing the grayscale range from overflowing the [0, 255] interval, to obtain the final expanded effective grayscale range.
[0017] Step Nine: Based on the actual effective grayscale range of each channel and the unified Gamma correction parameter, construct a piecewise pixel mapping model. For grayscale values below the lower limit of the actual effective grayscale range, grayscale suppression is used to map them to the lowest grayscale level, thereby suppressing noise in dark areas. For grayscale values above the upper limit of the actual effective grayscale range, grayscale enhancement is used to map them to the highest grayscale level, thereby enhancing the highlight of bright details. For grayscale values within the actual effective grayscale range, a nonlinear transformation mechanism is introduced, combined with the unified Gamma correction parameter, to adaptively adjust the grayscale distribution, thereby optimizing the stretching of the grayscale dynamic range and directional enhancement of image contrast.
[0018] Step 10: Based on the pixel mapping tables corresponding to the three channels mentioned above, perform mapping transformations on the pixels of each channel of the original image to obtain the enhanced pixel matrices of each channel. Merging these three matrices will yield the image img1.
[0019] Step 11: Enhance the saturation of the image by taking the maximum, minimum, and average values of the image img1 obtained in Step 10.
[0020] Step 12: Convert the obtained img1 image from the RGB color model to the HSV color model, extract the saturation component S information, and process the saturation S using an adaptive nonlinear method; this can effectively avoid over-enhancement of the image and color deviation; convert the HSV color space after nonlinear processing back to the RGB color space to obtain the final enhanced image.
[0021] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0022] The method of this invention adaptively determines the effective grayscale range and Gamma correction parameters by utilizing the histogram characteristics of each channel of the image. It requires no manual intervention, has strong scene adaptability, and can adapt to images with different grayscale distributions, effectively improving the contrast and detail of the image while avoiding color distortion. Attached Figure Description
[0023] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention;
[0024] Figure 2 A comparison diagram showing the effects of the original image and the image enhanced by the method of this invention;
[0025] Figure 3 This is a comparison diagram showing the effect of the image enhanced by the method of the present invention with images processed by other methods. Detailed Implementation
[0026] The technical solution of the present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.
[0027] See Figure 1 An adaptive image enhancement method for a television camera according to this embodiment includes the following steps:
[0028] S01. Image Reading and Channel Separation: The original color image is read using the OpenCV library, and the pixel matrix of the three channels (red (R), green (G), and blue (B)) is separated by the cv2.split function. The total number of pixels in the entire image, PixelAmount, is then calculated.
[0029] S02. Calculate the pixel histograms for the red (R), green (G), and blue (B) channels respectively. Each histogram contains the number of pixels corresponding to 256 gray levels (0-255), as shown in the following formula:
[0030]
[0031]
[0032] In the formula, channel is the channel currently being calculated, i and j are the row and column coordinates of the current pixel, and hist stores the number of times each pixel appears in each frame of the input image. hist is an array containing 256 data points.
[0033] S03. Effective grayscale range confirmation: Set LowCut and HighCut thresholds, traverse the histogram of each channel, accumulate the number of pixels in the histogram in the forward direction, and when the accumulated value reaches PixelAmount×LowCut, the corresponding value is the minimum non-zero grayscale value min; accumulate the number of pixels in the histogram in the reverse direction, and when the accumulated value reaches PixelAmount×HighCut, the value is the maximum non-zero grayscale value max.
[0034] S04. Calculate the extension amount delta, as shown in the following formula:
[0035]
[0036] Based on the obtained expansion amount delta, the maximum / minimum non-zero grayscale values are corrected respectively, as shown in the following formula:
[0037]
[0038]
[0039] S05. After calculating the mean brightness of each channel, calculate the corresponding Gamma parameter based on the mean brightness of each channel. Judge based on the mean brightness value. If the mean is less than or equal to 128, it indicates that the overall brightness of the original image is too low and the overall brightness of the image needs to be increased. The Gamma value is defined as follows:
[0040]
[0041] Conversely, if the mean is greater than 128, it indicates that the overall brightness of the original image is too bright, and the brightness of the image can be appropriately reduced. The Gamma value is defined as follows:
[0042]
[0043] S06. The average value of the Gamma parameters of the three channels is used to obtain avg_Gamma.
[0044] S07. Construct a mapping table using the maximum and minimum grayscale values calculated for each channel and avg_Gamma. Values less than the minimum are set to 0, and values greater than the maximum are set to 255. The mapping relationship between these values is as follows:
[0045]
[0046] S08. Traverse each pixel of the original image, transform the data of each channel according to the mapping table, and merge the transformed R, G, and B channels to obtain the enhanced image img1.
[0047] S09. Statistically compare the pixel values in the three channels of the enhanced image img1 to obtain the maximum value (max_bgr), minimum value (min_bgr), and mean value (mean_bgr) at each pixel position in the three channels.
[0048] S10. Convert the img1 image to the HSV color space, perform adaptive nonlinear processing on the S channel, and construct an adaptive adjustment factor. The specific formula is as follows:
[0049]
[0050] S11. Apply the adaptive adjustment factor k to the saturation component S in the HSV space to enhance saturation through a nonlinear transformation. The transformation formula is as follows:
[0051]
[0052] In the above formula The value is the processed saturation value, and the enhanced saturation value should be kept away from exceeding the effective range [0, 255].
[0053] S12, Enhanced saturation component The image is then merged with the hue (H) and lightness (V) and converted back to the BGR color space to obtain the final enhanced image.
[0054] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. An adaptive image enhancement method for a television camera, characterized in that, Includes the following steps: Step 1: Read the color image and separate the pixel matrix of the red, green, and blue channels; Step 2: Calculate the pixel histogram information for each channel and determine the grayscale range; Step 3: Determine the mean pixel value for each channel based on the pixel histogram, and calculate the Gamma correction parameter according to the formula to obtain the mean Gamma parameter. The Gamma correction formula is as follows: Where mean_intensity is the average brightness of the image, which falls within the range [0, 1]; Step 4: Based on the grayscale range and mean Gamma parameter, construct a pixel mapping table for each channel and perform pixel transformation to obtain the enhanced image img1; Step 5: Convert the enhanced image img1 to the HSV color space and perform adaptive nonlinear correction on the saturation component; Step 6: Merge the components and convert the image from Step 5 back to the RGB color space to obtain the final enhanced image.
2. The adaptive image enhancement method for a television camera according to claim 1, characterized in that, The grayscale range mentioned in step 2 is determined by the cumulative pixel distribution of the histogram, and the original grayscale range is expanded and corrected.
3. The adaptive image enhancement method for a television camera according to claim 1, characterized in that, The generation of the Gamma parameter in step 3 is adaptively adjusted based on the brightness characteristics of the pixel mean and constrained within a preset range.
4. The adaptive image enhancement method for a television camera according to claim 1, characterized in that, The saturation correction in step 5 is adjusted using an adaptive nonlinear function.