数学形态学在害虫种群密度监测中的应用

Monitoring population density of pests based on mathematical morphology

随着信息快速采集技术、计算机网络技术、人工智能、图像识别、决策支持系统等高新技术的发展,推动了"精细农作"技术体系的广泛实践。该文将数学形态学的基本运算方法及形态分水岭分割算法等图像处理信息技术运用到害虫种群密度监测中,根据昆虫飞行中CCD镜头区域远近及昆虫个体大小的先验知识,利用基于先验知识的流域分割算法,能有效地抑制背景及翅膀的影响,准确地识别出昆虫的个数,试验分析表明大大提高了害虫信息的采集效率及精度。

With the development of computer network technology, artificial intelligence, images recognition and decision support system, precision farming technology system is used more and more widely. In this paper, image information processing technology such as mathematical morphology and watershed segmentation algorithm is used to monitor pest population density. According to the distance between CCD and flying insects and size of insect individuals, watershed segmentation algorithm based on priori information is proposed, which can effectively restrain the impact of background and insects＇ wings, while at the same time raises the efficiency and precision of pest data acquisition greatly.