论文部分内容阅读
将图像压缩与ITS交通路况图像电传视讯相结合,提出了一种基于机器学习参数选择的多项式拟合图像压缩编码方法.我们定义了两类指标来度量数据扫描方法对多项式拟合方法的影响;还研究了用机器学习方法选择得到的参数对图像扫描数据进行单调化处理;进而研究了用多项式拟合预处理数据的方法进行图像数据压缩.该法简单方便、快速高效,并已针对小幅复杂交通路况图像在中低信噪比取得了良好的结果.该法还可以推广到感兴趣区位于图像中心的一类图像.
Combining image compression with ITS traffic image teletext video, a polynomial fitting image compression coding method based on machine learning parameter selection is proposed.We define two types of indexes to measure the effect of data scanning method on polynomial fitting method The parameters of machine learning method are also studied to monotonize the image scanning data. Then the method of polynomial fitting preprocessing data is studied to compress the image data. The method is simple and convenient, fast and efficient, Complicated traffic imagery achieves good results at low and medium SNR, which can also be extended to a class of images located in the center of the image.