Leaf image multi-feature integrated plant species identification method and device

Abstract

The invention relates to a leaf image multi-feature integrated plant species identification method and device. The method comprises the steps of obtaining plant leaf sample images of known species; obtaining binary images, leaf gray images and leaf texture images according to the leaf sample images; extracting leaf shape features and texture factures according to the leaf gray images, the binary images and the texture images; carrying out feature integration on the extracted leaf shape features and the texture factures through a main component analysis and linear judgment analysis combined method and carrying out dimension reduction on feature integration results, thereby obtaining training sample feature data; and training a support vector machine classifier according to the training sample feature data and corresponding plant species, thereby carrying out species identification on the leaf images of to-be-tested plants according to training results. According to the method and the device, image processing and machine learning principles are combined for plant species identification, and the plant classification and identification accuracy and efficiency are improved.

Description

technical field [0001] The invention relates to the technical field of computer image processing, in particular to a plant species identification method and device for multi-feature fusion of leaf images. Background technique [0002] There are about 400,000 species of plants on the earth, and my country is the world's second largest resource bank of plant species. The identification of plant species is the basic work of botany research and agricultural production. It is the basis for human beings to understand plants, master the evolution of plants, understand the basic characteristics of plants and make use of them. It can be applied to the identification of crop diseases and insect pests, crop growth monitoring and other fields. At the same time, it also has close relations with agriculture, forestry, animal husbandry, sideline, fishery, traditional Chinese medicine and other fields. [0003] Traditional plant taxonomy research generally requires manual measurement and d...

AI Technical Summary

Problems solved by technology

[0004] However, due to different growth environments and genetic differences, the leaves of different plants have large or small differences in shape and texture, and there is a lack of methods in the prior art that can accurately and quickly identify plant leaves. And then can't meet people's demand for identification, classification and retrieval of plant species

Images