Plant disease and insect pest identification method based on sparse network migration

Abstract

The invention discloses a plant disease and insect pest identification method based on sparse network migration, and belongs to the technical field of intelligent agricultural disease and insect pestidentification. The method comprises the following steps: designing a pruning algorithm, iteratively traversing a network, freezing redundant parameters in a source domain network, and generating a retrained optimal sparse sub-network structure; employing deep migration learning, migrating the sparse network to a target domain, proposing a sparse network migration hypothesis, verifying the feasibility of the sparse network, exploring the potential association between a target task and existing knowledge, and initializing the network through the weight of a source domain, and achieving the knowledge migration and reuse on the target domain; finally, finely adjusting the sub-network by using a small number of samples of the target domain data, optimizing the network performance, and finishing the task migration, thereby solving the practical application problem. Plant diseases and insect pests can be recognized, the network detection precision is improved through sparse migration, and meanwhile, the problems that a traditional deep method needs to train a dense network, calculation expenditure is large, the requirement for hardware is high, and popularization is not facilitated are solved.

Description

technical field [0001] The invention belongs to the technical field of detection of plant diseases and insect pests in intelligent planting. It is a method for identifying plant diseases and insect pests based on sparse network migration, which effectively solves the problems of dense network, high computing cost and high hardware requirements of the traditional deep learning model of plant diseases and insect pests. At the same time, the sparse structure is used Improve the accuracy of network detection of pests and diseases. Background technique [0002] The identification of plant diseases and insect pests is an important task in agricultural production, which requires timely and accurate detection of disease status, so as to take effective control measures. Traditional methods rely on experience for manual observation and identification, and the accuracy and efficiency are not satisfactory. Applying computer technology to timely and accurately identify pests and disease...

AI Technical Summary

Problems solved by technology

However, sample collection and labeling often require a large amount of overhead

[0006] (2) Training requires high hardware and high computational cost: To train a deep deep neural network, it is necessary to perform multiple trainings on tens of millions of parameters in the network. In each training, a large number of floating point numbers are required. operation

The computing power of the hardware device for training the network has high requirements, and even if it is trained on a high-performance computing device, model training still takes a lot of time

Images