基于改进的形态学二次重建蜜蜂蜂王模糊识别方法  

AN IMPROVED MORPHOLOGY-BASED SECONDARY RECONSTRUCTION METHOD FOR QUEEN BEE RECOGNITION

作　　者：白云飞 张翔[1] 林建[1] Bai Yunfei;Zhang Xiang;Lin Jian(College of Mechanical and Electrical Engineering,Fujian Agriculture and Forestry University,Fuzhou 350002,Fujian,China)

机构地区：[1]福建农林大学机电工程学院,福建福州350002

出　　处：《计算机应用与软件》2020年第11期139-145,共7页Computer Applications and Software

基　　金：福建省高原学科建设项目(712018014)。

摘　　要：采用机器视觉技术和模糊推理相结合的方法识别蜂王。根据巢脾上蜜蜂分布图像特点,使用改进的形态学二次重建方法,标记前景待识别蜂体,动态扫描图像提取蜂体长轴和短轴所在的离散点集。构建Mamdani型模糊推理系统,建立5条推理规则,得到待识别个体是蜂王的概率大小,选取不同的概率阈值,将蜂王从巢脾中初步识别出来。实验结果表明:该方法在两种不同环境条件下的识别成功率分别达到88.8%和84.6%,大大减轻了工作量,操作上具有一定的可行性。Machine vision and fuzzy inference are combined to identify the queen bee.According to the characteristics of honeybee distribution on the nest spleen,the morphological secondary reconstruction method was used to highlight foreground object.By using the dynamic scanning on the image,the discrete point sets of the long axis and the short axis were extracted respectively.The Mamdani fuzzy inference system was constructed,and five inference rules were established.To conclude the probability of the queen,we needed to select different probability threshold to recognize the queen bee from the nest spleen.The experimental results demonstrate that,under two different environmental conditions,the success rate of recognition is 88.8%and 84.6%respectively,which greatly reduces the workload and has certain feasibility in operation.

关 键 词：蜂王识别 二次重建 前景标记 动态扫描 模糊推理 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]