Leaf disease identification method and system

Abstract

The invention provides a leaf disease identification method and system. The method includes the following steps that: plaque detection is performed on a blade to be identified, each plaque region on the blade to be identified is obtained, the surface appearance feature words and color feature words of the same position in any plaque region form a word pair; the word pairs are converted into corresponding compound feature words in a compound feature dictionary one by one, and the composite feature histogram of the leaf to be identified is constructed according to the number of times of the appearance of the compound feature words on the blade to be identified; and the compound feature histogram is used as the input of a pre-obtained disease classifier, and the disease of the leaf is identified according to the output result of the disease classifier. With the leaf disease identification method and system of the invention adopted, a blade or disease plaque area is not required to be pre-segmented in an image, instead, the words in the compound feature vocabulary of the image are put into statistics, so that the type of the disease can be identified. The leaf disease identification method and system have the advantages of high recognition rate and high recognition speed.

Description

technical field [0001] The invention relates to the technical field of image processing, and more particularly, to a method and system for identifying leaf diseases. Background technique [0002] Traditional crop disease identification is performed by experienced experts. This identification method has low efficiency and high work intensity, and cannot make scientific and accurate diagnosis of diseases in real time. With the wide application of image processing technology and machine vision technology in the field of agriculture, automatic disease identification has become a trend in the development of intelligent agriculture. [0003] At present, most of the image recognition methods for crop diseases are manually collected diseased leaves and photographed under specific light and simple background, and then use a segmentation algorithm to segment the leaf target, and then need to further identify the diseased area on the segmented leaves. Segmentation, and finally extract...

AI Technical Summary

Problems solved by technology

The accuracy of this type of disease identification method is relatively high, but due to the manual control of environmental imaging, automatic disease analysis under real natural conditions cannot be realized

In addition, some studies have realized the identification of diseases under natural conditions, but it is still necessary to segment leaf regions and diseased parts in complex backgrounds in advance. real-time diagnosis

[0004] In disease identification, feature ex

Images