论文部分内容阅读
脐腐病是番茄栽培中严重的生理病害。本研究以机器视觉为主要技术手段,综合运用图像处理、色度学、模式识别等方面的知识,研究番茄脐腐病果的识别方法。首先采集、整理番茄脐腐病发病早期、中期和晚期的病果图像,进行预处理操作,包括图像增强方法和图像分割方法的研究,获得病态部位的图像;然后提取病态图像的颜色、形状、纹理等特征信息,找出有效的特征参数,建立特征数据库;利用模式识别技术实现番茄脐腐病的快速、准确、无损诊断。
Umbilicus rot is a serious physiological disease in tomato cultivation. In this study, machine vision as the main technical means, the integrated use of image processing, colorimetry, pattern recognition and other aspects of knowledge, the study of umbilicus rot fruit identification method. Firstly, we collected and sorted the disease images of early, middle and late stages of tomato umbilical rot disease, carried out pretreatment operations, including image enhancement methods and image segmentation methods, and obtained the images of pathological parts. Then we extracted the color, shape, Texture and other features of information to identify effective features parameters, the establishment of feature database; pattern recognition technology to achieve fast, accurate and non-destructive diagnosis of tomato umbilical rot disease.