利用计算机图像处理技术,对农作物病害的自动识别做了实验性研究。以黄瓜霜霉病为例,采集了黄瓜霜霉病的数字图像作为试验样本,进行了灰度化处理、二值化处理、平滑化处理和边缘检测处理等,从病斑的形状和颜色两方面提取特征参数,利用k近邻法对样本进行分类识别,识别率达到95%。 The use of computer image processing technology, the automatic identification of crop disease made an experimental study. Taking cucumber downy mildew as an example, the digital images of cucumber downy mildew were collected as test samples and processed by grayscale, binarization, smoothing and edge detection. From the shape and color of lesion The feature parameters were extracted, and the k nearest neighbor method was used to classify the samples. The recognition rate reached 95%.