基于主动光源透射率校正的水下图像恢复

Underwater image restoration based on new transmittance estimation

针对河流涵洞、水坝闸道等高浑浊度水体环境下,水下图像退化严重、模糊雾化的问题,提出一种基于主动光源透射率校正的水下图像恢复方法。首先,根据对水下成像的物理模型分析,通过缩小误差、逐步逼近的方法调整最佳校正参数,进而较为准确的估算求解出散射光透射率;其次,对主动光源直接照射的高亮区域的透射率进行校正补偿,消除"光斑效应";最后,利用水下成像模型计算得到目标反射光,完成图像恢复。实验结果表明,该算法能够有效恢复水下图像,为水利工程的检测维护、风险评估提供重要的基础。

Aiming at the problems of serious underwater image degradation and fuzzy fogging in high turbidity water environments such as river culverts and dam gates,an underwater image restoration method based on active light source transmittance correction is proposed.Firstof all,according to the physics of underwater imaging Model analysis,adjust the optimal correction parameters by narrowing the error and stepwise approximation,and then more accurately estimate and solve the scattered light transmittance;secondly,correct and compensate the transmittance of the bright area directly illuminated by the active light source to eliminate the"spot effect";Finally,Use the underwater imaging model to calculate the reflected light of the target and complete the image restoration.The experimental results show that the algorithm can effectively restore underwater images,providing an important basis for the inspection,maintenance,and risk assessment of water conservancy projects.