串行视觉通信下失真图像颜色校正方法仿真

发布时间：2018-06-02 23:51

本文选题：串行视觉通信 + 失真图像　；参考：《计算机仿真》2017年11期

【摘要】：颜色校正是提高图像视觉效果及内容感知理解程度的重要环节。因此,在串行视觉通信下,需要进行失真图像颜色校正方法研究。但是采用当前方法对失真图像进行颜色校正时,图像间的颜色亮度差异较大,存在校正结果易出现噪点及色块的问题。为此,提出一种基于梯度区域分割的串行视觉通信下失真图像颜色校正方法。上述方法先将输入图像帧分解为包含大尺度边缘信息的基图像和含有小尺度信息的细节层,获取失真图像颜色多尺度细节函数,从调色板中提取颜色特征信息,采用统计特征校正方法对颜色单一部分进行校正,采用改进的多维概率密度转换法对颜色复杂部分进行校正,从而完成对串行视觉通信下失真图像颜色的校正。仿真证明,所提方法颜色校正精度高,可以减少串行视觉通信下颜色失真,保持原图结构特征。
[Abstract]:Color correction is an important step to improve image visual effect and content perception. Therefore, in serial visual communication, color correction methods for distorted images need to be studied. However, when the current method is used to correct the color of distorted image, the difference of color brightness between images is large, and the problem of noise and color block is easy to appear in the result of correction. Therefore, a color correction method for distorted images based on gradient region segmentation in serial visual communication is proposed. The method firstly decomposes the input image frame into a base image containing large scale edge information and a detail layer containing small scale information, obtains the color multi-scale detail function of the distorted image, and extracts the color feature information from the palette. The single part of color is corrected by statistical feature correction method, and the complex part of color is corrected by improved multi-dimensional probability density conversion method, and the color correction of distorted image under serial visual communication is completed. Simulation results show that the proposed method has high color correction accuracy and can reduce color distortion in serial visual communication and maintain the original structure features.
【作者单位】：常州大学怀德学院;
【基金】：国家自然科学基金(61170121)
【分类号】：TP391.41

【相似文献】