基于人工鱼群算法最大熵多阈值成熟草莓图像分割.docVIP

基于人工鱼群算法最大熵多阈值成熟草莓图像分割 　　摘要：为降低算法的计算复杂度、加快搜索速度，将人工鱼群算法（AFSA）用于最大熵多阈值的快速寻优，提出了一种基于人工鱼群算法的最大熵多阈值的成熟草莓图像阈值分割算法（AFSAMEMT）。首先提取RGB彩色图像的R分量灰度图像及灰度图像信息，然后设计多阈值整数编码并将最大熵转化为编码的目标函数，最后利用AFSA寻优求得最大熵及其对应的阈值，进行图像分割。结果表明，AFSAMEMT相对OTSU等图像分割算法在复杂环境下不仅能达到更好的分割效果，而且有更好的分割效率。 　　关键词：图像分割；人工鱼群算法；多阈值；最大熵 　　中图分类号：TP274；TP391.4；S668.4 文献标识码：A 文章编号：0439-8114（2015）13-3265-04 　　DOI：10.14088/j.cnki.issn0439-8114.2015.13.051 　　Image Segmentation of Ripe Strawberry based on Maximum Entropy Multiple Threshold of Artificial Fish School Algorithm 　　QIN Lei1，2，RUAN Song1，ZHANG Wen-zhi1，LI Ya-qin1 　　（1. School of Mathematics and Computer， Wuhan Polytechnic University， Wuhan 430023， China；2. School of Automation， Huazhong University of Science and Technology， Wuhan 430074， China） 　　Abstract： Aiming to reduce the computational complexities and accelerate the search speed， the artificial fish algorithm（AFSA） was introduced to more rapid optimization of maximum entropy multiple threshold， a novel maximum entropy multiple threshold strawberry image segmentation method（AFSAMEMT） based on AFSA was proposed. Firstly， the R component gray images of RGB color images and their gray image information were extracted， and then the multiple threshold integer coding was designed and the maximum entropy was converted to the coding corresponding objective function， finally， the maximum entropy and their corresponding thresholds were obtained by AFSA and the images were segmented. The experimental results showed that AFSAMEMT algorithm in a variety of complex environments could not only achieve better segmentation effect than OTSU algorithm， etc， but also had better segmentation efficiency. 　　Key words： image segmentation；artificial fish school algorithm；multiple threshold；maximum entropy 　　草莓的栽培始于14世纪的欧洲，它的生态类型较多，不论是酷热的南部非洲，还是严寒的北美洲北部，世界上大部分国家均可以栽培草莓。目前世界草莓年产量已超过500万t，栽培面积超过30万hm2。2007年中国年产草莓已达130万t，居世界第一位，年草莓出口量约5万t。草莓成熟后必须及时采摘，目前草莓的采摘方式主要为人工采摘，草莓的日成熟量可达2 250～3 000 kg/hm2，两个劳动力每日需要10 h才能采摘完[1]。草莓采摘工作量的繁重已