Plant identification method and system based on convolutional neural network

Abstract

The invention discloses a plant identification method and system based on a convolutional neural network. The method comprises the following steps of collecting plant images, classifying and marking similar plants in the plant images, and obtaining a plant database; inputting the plant images in the plant database to the convolutional neural network, and training the convolutional neural network to get a feature matching model; receiving to-be-recognized images, extracting the image eigenvalues of the to-be-recognized images by using the feature matching model, calculating the similarity between the image eigenvalues and the plants in the plant database, and judging the classification of the plants which the to-be-recognized images belong to according to the similarity. The plant identification method and system based on the convolutional neural network achieve the adaptability of plant identification, reduce the design difficulty and calculation amount, and improve the efficiency of the plant identification.

Description

technical field [0001] The invention relates to the technical field of plant identification, in particular to a method and system for plant identification based on a convolutional neural network. Background technique [0002] In classical pattern recognition, the features of preset plant images (such as SIFT, HOG, LBP features) are generally extracted in advance. After extracting the features, encode the features, such as commonly used BoW, FisherVector, etc. Then put the features into a classifier, such as SVM, for 2 classifications, train the optimal classification surface, find the features that best represent a certain type of plant, and remove the features that are irrelevant and autocorrelated to the classification. However, the extraction of these features is too dependent on human experience and subjective consciousness, and the differences in the extracted features have a great impact on the classification performance, and even the order of the extracted features w...

AI Technical Summary

Problems solved by technology

However, the extraction of these features is too dependent on human experience and subjective consciousness. The difference in the extracted features has a great impact on the classification performance, and even the order of the extracted features will affect the final plant classification performance.

Images