自适应背景光估计与非局部先验的水下图像复原

有效地实现单幅水下降质图像复原对水下资源探索及环境监控领域的清晰图像获取具有极其重要的意义。为解决常用暗通道先验方法来复原图像时,背景光的估计易受白色物体干扰,且无法有效估计前景中白色物体透射率,复原质量不高的问题。本文提出了自适应背景光估计与非局部先验的水下图像复原算法。首先根据背景光具有高亮度及平坦性的特点,利用阈值分割算法获得背景光的候选区域,再通过图像的色调信息从候选信息中选取最佳的背景光点。随后,利用各颜色通道光的波长与散射系数的相关性,提出了适用于水下图像的非局部先验,并利用该先验估计各通道的透射率。最后针对复原结果中,因水下介质,微生物,水流影响而产生的加性噪声,设计去噪的最小优化问题,并利用引导滤波求解该问题,以去除复原结果中的加性噪声。实验表明:该算法在确保运行效率的基础上,准确地估计透射率,较常用算法的复原精度提高了约18%。证明了该算法能有效用于单幅水下图像复原的工程实践中。

Underwater Image Restoration With Adaptive Background Light Estimation and Non-local Prior

Objective: It is significant to realize effective single underwater image restoration for acquiring clear image in underwater exploration and underwater environment monitoring field. Method: Most of existing algorithms used dark channel priors to restore image, which leads to misestimating the background light and transmission map. Hence, a novel method with adaptive background light estimation and non-local prior is proposed. Firstly, the candidate water light regions can be obtained by threshold segmentation algorithm due to the fact that water light regions have the properties of flat and high brightness. Then the water light value can be decided from the candidate regions by the dominant tone of input image. Secondly, the non-local prior is built to estimate the transmission map through taking into account the wavelength dependency of the attenuation. Finally, for removing the additive noise from medium and microorganism, a minimal optimization problem with the solution strategy of guided filter is proposed for obtaining the de-noising result. Result: Experimental results verify that the proposed algorithm can not only ensure the running efficiency, but can also estimate correct transmission map. Generally speaking, the restoration precision has improved 19% comparing with the existing algorithm. Conclusion: It can be used in the engineering practice of restoration of single underwater image.