Online semi-supervised learning applied to an automated insect pest monitoring system
| Autor: | Jui-Yung Chung, Ju-Chun Hsu, Dan Jeric Arcega Rustia, Ta-Te Lin, Ya-Fang Wu, Jun-Jee Chao, Chen-Yi Lu |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Online model
Computer science Soil Science Semi-supervised learning Machine learning computer.software_genre 01 natural sciences Classifier (linguistics) Contextual image classification Artificial neural network business.industry Deep learning 010401 analytical chemistry Supervised learning 04 agricultural and veterinary sciences Automation 0104 chemical sciences ComputingMethodologies_PATTERNRECOGNITION Control and Systems Engineering 040103 agronomy & agriculture 0401 agriculture forestry and fisheries Artificial intelligence business Agronomy and Crop Science computer Food Science |
| Zdroj: | Biosystems Engineering. 208:28-44 |
| ISSN: | 1537-5110 |
| DOI: | 10.1016/j.biosystemseng.2021.05.006 |
| Popis: | The unavailability and variability of training samples are the two essential concerns in the training of deep neural network models for image classification. For automated image monitoring systems, these problems are difficult when training a model through supervised learning methods because of the time and effort required. This paper proposes an adaptive solution to this problem by applying online semi-supervised learning to an automated insect pest monitoring system. The method used includes unsupervised pseudo-labelling of insect images and the training of semi-supervised classifier models for insect image recognition. The pseudo-labelling algorithm includes three major components: image labelling, label reconfirmation, and sample cleaning. Experiments were conducted on two unlabelled 1-year insect image datasets to evaluate the efficacy of the proposed method. It was found that the pseudo-labelling algorithm could achieve accuracy up to 0.963, hence enabling automated training data collection. The temporal improvement of the insect recognition performance by including new training data to retrain the classifier model was comparable in performance to the supervised learning approach as evaluated by cluster density, silhouette score, and F1-score. The proposed method was also able to automatically collect quality samples and train models regardless of the complexity of the images, making it a good alternative to replace laborious supervised learning. The proposed method can prevent contamination of a training dataset when images from new locations are collected. The presented techniques may also be used in other continuous learning applications that require automated training data collection and online model update. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect