摘要图像分割是数字图像处理的基础， 如何准确快速分割图像是研究的特点和难点之一。在图像分割算法中，阈值分割方法属于自动分割，不需要人工参与，得到广泛的应用和发展，其中三文 Otsu 算法在求取阈值时充分考虑图像的灰度信息、空间信息和像素点信息，分割效果较好，但由于涉及到图像信息量多，增加了算法的复杂度，降低了运算速度。为了解决这一问题，通过分析三文 Otsu 算法的基本原理，结合三文 Otsu 算法和遗传算法的优点，本文提出用遗传算法优化三文 Otsu 算法，即为了求取目标函数的最大值，通过遗传算法的编码、初始化、确定适应度函数以及选择、遗传和变异等一系列的运算的过程来求得最佳阈值。当目标函数最小时，对应的阈值是最优的阈值，根据阈值对图像进行分割。为了进一步说明优化算法的有效性，将其与三文 Otsu 算法分别从算法的速度和性能上进行仿真对比实验，分割效果表明本文方法在分割效果不变的情况下速度更快。39656
毕业论文关键词： 图像分割 阈值分割 OTSU 遗传算法 阈值 AbstractImage segmentation is the basis of digital image processing, and how tosegment the image accurately and quickly is one of the features anddifficulties of the research.. In the image segmentation algorithm, thresholdsegmentation method, which belongs to the automatic segmentation, withouthuman involvement, get extensive application and development, which threedimensional Otsu algorithm in obtaining thresholds fully consider the grayinformation of the image, space information and the pixel information, a goodsegmentation effect, but due to the amount of image information, an increaseof algorithm complexity. To reduce the computing speed.In order to solve this problem, through the analysis of the basic principleof three dimensional Otsu algorithm, combined with the advantages of threedimensional Otsu algorithm and genetic algorithm is proposed in this paper isto optimize the three-dimensional Otsu algorithm and genetic algorithm, inorder to obtain the minimum value of the objective function, through geneticalgorithm encoding, initialization, determining the adaptation degree functionand a series of operation process to obtain the optimal threshold. When thetarget function is the minimum, the threshold is the optimal threshold, and theimage is segmented according to the threshold..In order to further illustrate the effectiveness of the optimizationalgorithm, and the three-dimensional Otsu algorithm when compared to asimulation experiments are done from the speed and performance of the algorithm, the segmentation results that this method has better segmentationeffect, the advantages of faster and better than the three dimension Otsualgorithm.
Keywords:Image segmentation Threshold segmentation OTSUGenetic algorithm Threshold
目录
摘要. I
Abstract..II
1 绪论. 1
1.1 研究背景. 1
1.2 国内外研究状况3
1.2.1 阈值分割算法.3
1.2.2 基于边缘的分割方法..4
1.2.3 基于区域的分割方法..5
1.3 论文组织结构. 5
2 阈值分割、OTSU算法 7
2 .1 图像分割的基础知识. 7
1.3.1 图像分割的基础知识..7
1.3.2 图像分割的应用领域..8
2 .2 阈值分割 9
2.2.1 阈值分割的基础知识9
2.2.2 经典的阈值分割算法.11
2.3 最大类间方差算法 14
2.3.1 最大类间方差算法的基本知识 14
2.3.2 经典的 Otsu算法模型.. 15
2.3.3 三文 Otsu阈值分割法.. 16
2.3.4 实验仿真及分析. 20
2.4 本章小结.21
3 基于遗传算法的三文Otsu算法优化22
3.1 遗传算法的基础知识..22
3.2 优化目的.24
3.3 基本流程.25
3.4 实现方法.27
3.5 仿真实验及分析.29
3.6 本章小结.. 31
4 总结.32
致谢 33 阈值分割的研究+文献综述:http://www.751com.cn/zidonghua/lunwen_40038.html