基于同态滤波的水下图像增强与色彩校正模型

Underwater image enhancement and color correction model based on homomorphic filter

水体对于不同波长的光信号衰减程度不一致,这种现象破坏了水下图像的清晰度和色彩恒定性。为了解决水下图像亮度与色彩扭曲问题,提出一种基于同态滤波的水下图像增强与色彩校正模型。首先,通过比尔-朗伯定律和路径辐射分量构建出水下成像模型。其次,通过同态滤波对未经过衰减的水下图像进行估计。最后,通过麦克劳林级数对水下成像模型进行级数展开,进而推导出一种保持颜色恒定的水下图像色彩校正模型。实验部分分别对比了水下图像的主观视觉效果和客观评价指标,验证了该算法能够有效地保证水下图像的清晰度和色彩恒定性。校正后的水下图像细节丰富,色彩逼真。

The degree of light attenuation varies corresponds to the different wavelengths in water substance. The clarity and color constancy for underwater image are restricted due to this property. An underwater image enhancement and color correction model based on homomorphic filter is proposed to deal with the luminance and color distortion for underwater image. Firstly, the underwater imaging model is established by Beer-Lambert law. Secondly, homomorphic filter is employed to estimate the underwater image without light attenuation. Finally, an underwater image color correction model is derived from the series expansion of underwater imaging model to keep color constant. Experiments compare the subjective visual effect and objective assessment index of underwater image. Experimental results show that the proposed model is effective to enhance the clarity and color constant for underwater image. The proposed model produces a detailed and real colorful underwater image after correction.