基于 UCM 算法中直方图拉伸的水下图像增强研究

针对水下图像色彩失真,对比度低以及图像模糊不清等一系列问题,提出了一种基于色彩模型的水下图像增强算法,改进了现有的 UCM 算法的直方图线性拉伸,使用限制对比度自适应的方法对直方图进行非线性均衡化,使处理后的图片对比度增强效果更好,图像质量更高,更符合人类的视觉感知。 通过搭建的水下海参场景和模拟海底图像偏蓝绿色的实验环境,拍摄了 4 组海参在不同姿态、工况下的水下图片并对其进行定性、定量分析,得出改进后的 UCM 算法在 UIConM、UIQM、NIQE 指标下的数值分别平均为 0. 63、4. 30、3. 30,相比较其他 3 种算法,该算法处理后的图像质量评估指标均为最优,由此证明了研究的算法相对于其他的传统算法显示出更好的可行性和优越性,并且能够适应不同的水下工况,拥有良好的鲁棒性。

Research on Underwater Image Enhancement Based on Histogram Stretching in UCM Algorithm

Aiming at a series of problems such as color distortion, low contrast, and blurred image of underwater images, this paper proposes an underwater image enhancement algorithm based on the color model,which improves the histogram linear stretching of the existing UCM algorithm, and uses the contrast limitation adaptive method to carry out nonlinear equalization of the histogram. Images processed with the proposed algorithm show a better image contrast enhancement effect, have better quality and are more in line with human visual perception. Through the construction of the underwater sea cucumber scene to simulate the experimental environment of the blue-green seabed image, 4 groups of sea cucumbers under different poses and workingconditions were photographed for experiments, qualitative and quantitative analysis. It is concluded that the average values of the improved UCM algorithm under UIConM, UIQM and NIQE indexes are about 0. 63, 4. 30 and 3. 30 respectively. Compared with the other three algorithms, the image quality evaluation indexes processed by this algorithm are all optimal, which proves that the proposed algorithm has better feasibility and superiority compared with other traditional algorithms, can adapt to different underwater conditions, and has good robustness.