一种叶类蔬菜病害诊断方法及系统

本发明提供一种叶类蔬菜病害诊断方法及系统，涉及蔬菜病害诊断技术领域，具体步骤包括：S1.采集蔬菜的叶片图像，并对所述蔬菜的叶片图像进行图像缩放，将叶片图像的尺寸统一调整为224*224大小，并将叶片图像复制成两组相同的图像，一组进行灰度化处理，获得第一识别图像。本发明相较于计算所有图片的纹理特征值的均值作为阈值点，更加局部化，能够降低异常值的影响，通过针对性地考虑健康状态下叶片的特征，确定阈值，使得感染区域的变化更加准确地反映出病害的发展情况，有助于提高诊断的准确性，减少误判的可能性，将健康区域蔬菜叶片的颜色和纹理特征作为阈值，能够更准确地反映出健康状态下叶片的特征分布。 The invention provides a leaf vegetable disease diagnosis method and system, and relates to the technical field of vegetable disease diagnosis, and the method specifically comprises the steps: S1, collecting a leaf image of a vegetable, carrying out the image zooming of the leaf image of the vegetable, uniformly adjusting the size of the leaf image to 224 * 224, copying the leaf image into two groups of same images, and carrying out the image zooming of the leaf image of the vegetable; one group is subjected to graying processing to obtain a first recognition image; compared with calculation of a mean value of texture feature values of all pictures as a threshold point, the method is more localized, can reduce the influence