Agriculture has been the basic industry of mankind since ancient times. Accurate and robust crop pest identification system is an important step for reliable prediction of agricultural pests in precision agriculture community, which has attracted great attention from many countries. The development of computer vision and deep learning injects new vitality into this field. Many intelligent methods have been introduced into the field of pest recognition. However, long-time model preparation is the bottleneck of these methods, and expensive labels also make this method limited. At the same time, they can not be well compatible and extended when new pests appear. Therefore, we developed a pest identification algorithm based on self supervised learning. In the self supervised learning model(FE-VIT) proposed in this paper, we train a pretraining model through unlabeled data. In the pretext task, the purpose is to make the model match the downstream task(pest recognition). Besides, we simplify the model structure, reduce the time complexity from O (N2) to o (n*k), and make the proposed model Extensible. Our method can achieve about 17% time reduction and 1.1% accuracy improvement. A large number of detailed experiments show the accuracy and reliability of our model. Our code will be released in https://github.com/54zanly/FE-VIT .
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.