农田覆膜播种中白色污染图像盲复原方法研究

Study on Blind Restoration Method of White Pollution Image in Farmland Mulching

图像盲复原能够恢复图像里有价值的信息并改善图像质量,因此该课题在图像处理领域具有较高的研究性和实用价值。在农田覆膜播种白色污染图像恢复中,该技术的应用可获取高质量图像,为白色污染治理提供依据,具有一定现实意义。首先,因为稀疏特性表现在白色污染自然图像边缘,可以利用一种权重的全变差范数对图像实现正则化约束,考虑到运动模糊点扩散函数的特点,把运动模糊函数的连续平滑性及稀疏性运用到图像正则化约束分析中来,达到白色污染自然图像边缘稀疏和锐化增强的效果,这样对图像复原更为有利;其次,以图像正则化约束分析为基础,依据图像先验信息和正则化特点,利用改进版Bregman迭代法建立图像盲复原代数函数,为简化求解在函数中加入惩罚和分裂因子,可以有效获取高质量的白色污染盲复原图像。仿真实验证明,运用本文所述方法能获得更高质量的农田覆膜播种中白色污染图像,为治理污染提供信息资料。

Image blind restoration can restore valuable information and improve image quality,so the subject has high research and practical value in the field of image processing.The application of this technology in the restoration of white pollution image in plastic film mulching can obtain high quality image and provide the basis for white pollution control.It has certain practical significance.Firstly,the sparseness property is expressed in the edge of the natural image.The regularization constraint can be achieved by using the total variance norm of one weight.Considering the characteristics of the motion blur point diffusion function,the smoothness of the motion blur function and Secondly,on the basis of the image regularization constraint analysis and based on the priori information of the image,this paper proposes a new algorithm based on the image regularization constraint analysis and the image regularization constraint analysis.Secondly,the image regularization constraint analysis is applied to the image regularization constraint analysis,and the sparse edge image is sharpened and sharpened.And the regularization feature,the improved Bregman iterative method is used to establish the blind restoration algebraic function of the image.In order to simplify the solution,the penalty and the splitting factor can be added to the function to obtain the high quality white image.The simulation results show that the method described in this paper can obtain a higher quality image of white pollution in plastic film sowing,and provide information for controlling pollution.