基于MATLAB的直方图均衡算法研究与实现.doc

摘 要 基于灰度直方图的图像增强技术是图像增强领域的一个重要方面。直方图均衡化处理是增强图像对比度的方法之一，其方法是通过使用累积函数对灰度值进行调整以实现对比度的增强。从本质上说，直方图均衡化处理是把原始图像的灰度直方图从比较集中的某个灰度区间变成在全部灰度范围内的均匀分布，其主要缺点是处理后的图像将会大量损失灰度层次，而且增强力度不够。 本文首先介绍了MATLAB的相关概况和技术特点。接着，提出了直方图均衡化的图像增强原理。然后，提出了直方图均衡化的经典算法和两种改进算法。最后，以两幅医学图像为例，对文中提出的直方图均衡化算法进行了MATLAB仿真，并对仿真结果进行了比较和分析。 MATLAB仿真结果表明：直方图均衡化使得图像的对比度增强，许多细节显得更加清晰，有利于图像分析与识别。不同于理论值，均衡化后的直方图并不是完全均匀分布的，而是近似均匀分布的。此外，直方图均衡化对于对比度较弱的灰度图像增强效果比较明显。最重要的是，相对于经典直方图均衡化算法，两种改进算法能够更加有效地改善灰度图像的对比度和灰度动态范围。 关键词：直方图均衡；图像增强；MATLAB仿真；医学图像ABSTRACT Histogram-based image enhancement technology is an important aspect in the field of image enhancement. Histogram equalization is one of the ways to enhance contrast. It can enhance contrast by accumulative function to adjust the gray value. In essence, histogram equalization can transform the histogram of the original image that is relatively concentrated in certain regions into uniform distribution in the whole range of gray-level. The main drawback of histogram equalization is that the image after being processed will lose a great amount of gray levels and the enhancement effect is not so obvious. Firstly, relevant overview and technical characteristics of MATLAB is introduced in this paper. Secondly, image enhancement principle of histogram equalization is advanced. Thirdly, classical algorithm and two modified algorithms of histogram equalization are advanced. Finally, histogram equalization algorithms being advanced in this paper are simulated in MATLAB based on two medical images and then the results of MATLAB simulation is compared and analyzed. The results of MATLAB simulation demonstrate that histogram equalization can enhance contrast to make details more clear which is benefit for image analysis and identification. Histogram of equalization image is not entirely uniform distribution but approximately uniform distribution is different with theory. Additionally, weak contrast gray image can acquire relatively obvious enhancement effect by histogram equalization.