基于光学模型的退化叶片图像复原方法

Degraded leaf image restoration approach based on optical imaging model

针对空气浮尘造成植物叶片图像颜色失真、细节信息模糊问题,采用一种基于光学成像模型的灰尘去除算法,对退化叶片图像颜色进行恢复。首先根据灰尘颗粒对光线的反射和散射特性,建立叶片图像退化模型;然后利用有尘黑卡对光谱的反射特性计算出灰尘的反射率,在此基础上结合暗元原理和小波变换估计出入射光强及透射率2个模型参数。采用色卡和葡萄叶片图像测试算法的有效性。试验结果表明:本研究算法对叶片颜色有较好的恢复效果,复原后图像a、b分量(CIE Lab颜色模型)的平均绝对误差显著降低,有效地提高了图像的对比度和清晰度。试验还表明,本研究算法对不同天气和照明条件、不同品种的葡萄叶片图像均有较好的颜色恢复效果。

In order to tackle the problem of color distortion caused by atmospheric dust,a method based on optical imaging model is developed to restore the color of leaf image.The model considers the optical characteristics of reflection and scattering of a dust layer.The parameters of model are estimated as follows：The dust reflectance is calculated according to the reflection characteristics of black part of dusty ColorChecker Chart,and the incident optical intensity and transmittance are estimated by dark channel prior and wavelet transforms.The images of dusty ColorChecker Chart and grape leaves are employed to check the effectiveness of the algorithm.The results show that the method can significantly reduce the mean absolute error of aand b color components of CIE Lab color model of the degraded images,effectively enhance their contrast and quality.The results also indicate that the method is applicable to images under various conditions such as weather,illumination and grape varieties.