TSPL插值算法在图像放大方面的应用
时间:2019-05-03 08:37 来源:毕业论文 作者:毕业论文 点击:次
| 摘要随着时代的发展，图像在人们生活中应用得越来越广泛，人们对图像质量的要求也越来越高。如何从有限像素的低分辨率图像得到高分辨率图像的研究变得很有价值，例如这些技术在军事、航空、医疗等方面的应用。在图像放大方面，现有的经典插值算法仍存在着很多的问题，如经过插值放大后得到的图像边缘会出现马赛克现象，图像变得模糊等等。35099 经典的插值算法的优势在于计算方便，但现如今对图像质量的要求越来越高，原有的经典插值算法已不能够完全满足需求，所以新的插值方法也层出不穷，如B-样条插值以及一些改进的经典插值算法等等。 本文实现了三种经典的插值算法，并且实现了《基于三子集划分的灰度图像分层插值新算法》中提到的基于TSPL的图像插值放大算法。该文中提到了一种新的插值算法，首先将一幅图像分成若干基函数之和，然后对各个基函数进行放大，再将这些放大后的基函数相加组合成一幅新的图像，这幅新的图像就是放大后的图像。我们通过实例和主客观的分析，对几种经典插值算法基于TSPL的图像插值放大算法进行比较，经验证新的插值算法放大后得到的图像质量要好很多。 毕业论文关键字：插值；图像放大；双线性插值；TSPL插值算法；matlab图像处理； Abstract With the development of the times, the application of image in people’s life is getting wider and wider and people’s demand of the quality of image is higher and higher. The research on how to get high-resolution image from low-resolution image with limited pixels becomes valuable such as the technology used in the aspect of military, aviation and medical and so on. In the aspect of image enlargement, there are still lots of problems in the existing classical interpolation algorithms such as there will be mosaic phenomenon at the edge of the image after interpolation and the image may become fuzzy too. The advantage of classical interpolation lies in computing convenience, but nowadays, people’s demand of the quality of image getting higher and higher, the original classical interpolation algorithm can’t meet the needs completely. So the new interpolation methods such as B-spline interpolation and some improved classical interpolation algorithm and so on come out. Three classical interpolation algorithms are realized in this paper. And a new interpolation algorithm which proposed by a new algorithm for interpolation from images based on TSPL is realized too. In this paper, a new interpolation algorithm is proposed, firstly, we pide an image into a number of basis of functions, secondly, we enlarge the various basis functions one by one, then add the sum of these enlarged basis functions into a new image, and the new image is the enlarged image. Through the examples and the analysis of subjective and objective view comparison of several classical interpolations and the interpolations based on TSPL, we found that the quality of the image enlarged by the new interpolation is better. Keywords: interpolation; image enlargement; bilinear interpolation; TSPL interpolation algorithm; matlab image processing; 目录 1.绪论 1 1.1 研究背景、目的和意义 1 1.2 图像插值放大技术的国内外研究现状 1 1.3 图像插值放大技术的应用与发展趋势 2 1.4 论文内容 2 1.5 论文结构 3 2.三种经典图像插值方法 5 2.1 最近邻域插值法 5 2.2 双线性插值法 6 2.3 双三次插值法 7 3.基于TSPL的图像放大算法 9 3.1 TSPL算法描述【5】 9 (责任编辑：qin) |