基于计算机视觉的鞘翅目害虫自动鉴定

鞘翅目害虫的鉴别工作是由数量有限的专业研究人员进行，大部分非专业人员因缺乏相关知识和技术培训，要快速鉴定害虫种属具有相当的难度，不利于害虫防治。为此提出利用计算机视觉技术对鞘翅目害虫进行自动识别。用不同的图像获取系统分别采集体长不同的鞘翅目害虫图像，然后通过数字图像分析技术，提取和计算害虫图像的几何形状特征参数。采用模糊模式识别算法建立害虫的模式库和隶属函数，然后根据模糊集理论的极小极大准则完成待识别害虫的归类。对29 种鞘翅目害虫进行识别，识别正确率为
86.2%。为鞘翅目害虫的快速鉴定提供了一个新的途径。
关键词：鞘翅目害虫; 计算机视觉; 特征提取; 自动识别
Abstract: Identification of Beetle Insects has been performed by human experts with rich experience. However, sufficient number of competent human experts are not available to cover the large area. This paper presents a computer vision-based method to identify beetle insects. Two image acquisition systems were designed to capture the images with respect to insects’ length. And median filtering was used to enhance the images for further segmentation. Parameters related to shape features were extracted to represent the image content. Fuzzy classifier was adopted to identify beetle insects. Results showed that the correct recognition rate is 86.2% . It is a feasible way for beetle insect’s rapid identification.
Key words: Beetle Insect; Computer Vision; Feature Measuring; Automated Identification