计算机视觉技术在植物病害识别上的研究进展

Advances in recognition of plant diseases based on computer vision

随着农业和现代化信息技术的交互、联结和碰撞,农业逐渐趋于现代化、智能化和数字化,近年来运用计算机视觉技术对植物病害进行诊断得到广泛应用,比传统方法更加迅捷、精确。分别从图像采集、图像预处理、图像分割、图像特征提取、病害识别和分类5个方面进行阐述,总结了植物病害图像识别技术的要点及存在问题,并对其未来发展进行了展望,为计算机视觉技术在植物病害识别上的应用和研究提供依据。

Plant diseases restrict the development of agriculture in production,safety and economy.Monitoring plant health status and preventing plant diseases from occurrence are very important for sustainable agricultural development.With the interaction,connection and collision between agriculture and modern information technology,agriculture gradually tends to be modern,intelligent and digital.Computer vision has been widely used in detecting plant diseases in recent years,which is more rapid and accurate than traditional methods.The computer vision in recognition of plant diseases was reviewed from the aspects of image acquisition,image preprocessing,image segmentation,image feature extraction and disease recognition and classification,and its key points were summarized.Problems arising from and outlook on the image recognition of plant diseases based on computer vision were put forward to provide some reference for the application and research of computer vision in recognition of plant diseases in the future.