激光与光电子学进展, 2020, 57 (12): 121016, 网络出版: 2020-06-03
基于白平衡和相对全变分的低照度水下图像增强 下载: 1150次
Low-Illumination Underwater Image Enhancement Based on White Balance and Relative Total Variation
图像处理 图像增强 光照补偿 白平衡 引导滤波 相对全变分 image processing image enhancement illumination compensation white balance guided filtering relative total variation
摘要
针对低照度水下图像存在色偏和细节模糊的问题,提出一种基于白平衡和相对全变分的低照度水下图像增强算法。该算法依据光线在水中选择性衰减的特性,对水下图像进行全局光照补偿以提高图像亮度,并采用灰度世界算法校正水下图像颜色;依据引导滤波的保边平滑性构造新的相对全变分约束来估计照度图,并基于低照度成像模型得到反射图像,即增强后的水下图像。主观效果、客观分析及应用比较实验均证明了本文算法在校正色偏和增强图像细节方面的有效性。
Abstract
To solve the problem of color cast and detail blur in low-illumination underwater images, a low-illumination underwater image enhancement algorithm based on white balance and relative total variation is proposed. According to the characteristics of selective attenuation of light in water, the algorithm firstly performs global illumination compensation for underwater images to improve image brightness and uses gray world algorithm to correct underwater image color. Then a new relative total variation constraint is constructed according to the edge preserving smoothness of guided filtering to estimate the illumination map. Finally, a reflection image, namely an enhanced underwater image, is obtained based on the low-illumination imaging model. The subjective effect, objective analysis, and application comparison experiments all prove the effectiveness of the proposed algorithm in correcting color cast and enhancing image details.
张薇, 郭继昌. 基于白平衡和相对全变分的低照度水下图像增强[J]. 激光与光电子学进展, 2020, 57(12): 121016. Wei Zhang, Jichang Guo. Low-Illumination Underwater Image Enhancement Based on White Balance and Relative Total Variation[J]. Laser & Optoelectronics Progress, 2020, 57(12): 121016.