可变光照条件下多维图像灰度校正方法仿真

摘要

A gamma correction method of multi-dimensional image under variable illumination condition is proposed based on the homomorphic filtering.Firstly,the image picture frequency filtering with the grey level transformation process integrated with the homomorphic filtering theory is combined,and the illumination/reflectivity model of multi-dimensional image is used as the basis of image frequency domain processing.Then,the multi-dimensional image quality is improved under variable illumination condition integrated with the compression grey range of multi-dimensional image and enhancing contract,and the multi-dimensional image after homomorphic filtering process is obtained.The point belonging to the background form the original multi-dimensional image is directly selected by using the multiplying factor theory according to the multi-dimensional image feature.Moreover,the entire multi-dimensional background image under variable illumination condition is obtained through interpolation integrated with the surface fitting.Thus,the gamma correction of multi-dimensional image under variable illumination condition is completed.The simulation results show that the method improves the pretreatment effect of multi-dimensional image apparently compared with current method.It can reduce the illumination uneven phenomenon of image effectively and increase the robustness of image gamma correction.%针对当前方法在对可变光照条件下多维图像灰度校正处理时容易造成边缘模糊,局部对比度得不到增强的问题,提出一种基于同态滤波的可变光照条件下多维图像灰度校正方法.该方法先结合同态滤波理论将图像频率过滤和灰度变换过程相结合,依据多维图像的照度/反射率模型作为图像频域处理的基础,结合多维图像压缩灰度范围和增强对比度来改善可变光照条件下多维图像的质量,获得经过同态滤波处理后的多维图像,在此基础上利用乘系数理论依据可变光照条件下多维图像特性,直接从原多维图像中选取出属于背景上的点,通过插值结合曲面拟合得到整个可变光照条件下多维背景图像,由此完成对可变光照条件下多维图像灰度校正.实验结果表明,所提方法与当前方法相比,多维图像的预处理效果明显增强,能够有效地减少图像的光照不均匀现象,图像灰度校正的鲁棒性较优.