Crop leaf type recognition method based on multi-view multi-task ensemble learning

Abstract

The invention relates to a crop leaf type recognition method based on multi-view multi-task ensemble learning. The method comprises the steps of selecting a leaf image as an original data set, carrying out the feature extraction, and obtaining a data set under a plurality of views; using a CNN model as a base learner, and carrying out independent ensemble learning on the data sets under the viewsand the original data set; fixing parameters of all the base learners, removing the last layer of the full-connection classifier in the base learners, splicing outputs of all the models, adding a newclassifier, and performing joint feature selection on a plurality of views to enable the accuracy of a verification set to reach expectation so as to obtain models under the plurality of views; and recognizing the types of the leaves by utilizing multi-task learning. According to the method, the accuracy and generalization ability of the model are enhanced, and the problem of weak generalization ability caused by insufficient training data of a traditional deep learning model and simple stacking depth of the model is solved on the whole.

Description

technical field [0001] The invention belongs to the field of artificial intelligence, and proposes an improved method for improving the effect of identifying crop leaves and existing diseases on the basis of a traditional deep learning model. Background technique [0002] At present, the problem of food security is becoming more and more serious. There are many factors that are threatening food security, among which plant diseases pose a serious threat to food security on a global scale. In the past, the identification of crop diseases was mostly done manually, but there are many deficiencies in manual identification. With the rise of precision agriculture, the use of information technology to assist agricultural production provides new ideas for the identification of crop diseases. Image processing technology is one of them. It has various advantages that traditional methods do not have for crop disease identification, namely real-time Strong, fast, low misjudgment rate, ...

AI Technical Summary

Problems solved by technology

[0007] At present, although the image recognition method based on deep learning is better than many traditional algorithms in terms of accuracy, a large number of crop recognition models still face the problem of generalization ability of the model. There are two main reasons for this: First, the size of the data set limits

The manual acquisition and manual labeling of crop disease pictures is time-consuming and labor-intensive, resulting in less data available for model training. The traditional solution is to use data augmentation to expand the data set, but the degree of improving the generalization ability of the model is very limited

The second is that a large number of crop recognition models simply use a simple stack of convolutional neural networks and fully connected layers, and they are all implemented under the idea of ​​​​single view. However, different views of objects describe different characteristics of objects. Defects lead to poor generalization ability of the model itself

Images