Semantic Segmentation of Rice Disease Images based on DeepLabV3+

DOI

10.2991/978-94-6463-300-9_95How to use a DOI?

Keywords

DeepLabV3+; semantic segmentation; rice leaf disease; ASPP; intelligent agriculture

Abstract

Disease area segmentation is an important task in the field of smart agriculture, which is of great significance for analyzing the fine-grained information inside disease spots and supporting prevention and control decisions. Early disease area segmentation mostly relied on image processing or manual features, and its accuracy could not meet the practical application requirements in field scenarios. Thanks to the rapid development of pattern recognition technology, semantic segmentation algorithms based on deep learning provide new solutions for accurately and automatically identifying diseased areas. In this paper, we present a semantic segmentation approach for rice leaf disease images using DeepLabV3+. Specifically, we combine the encoder-decoder structure with atrus convolution as well as the spatial pyramid pooling to further improve segmentation accuracy. We constructed a dataset of rice leaf images containing four different types of diseases, and trained and tested models on this dataset. The model performance is evaluated with standard metrics such as mean intersection over union (mIoU) and pixel accuracy. In addition, we design some other sets of corresponding experiments to test the performance in some specific circumstances, including in poor light conditions, on the background of different situations, with low resolution and with noises. All outcomes demonstrate the efficacy and reliability of our methodology. We also discuss the challenges and limitations of the model, as well as possible future directions for improvement.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Jinghao Li
AU  - Zirui Ren
AU  - Letian Zhou
PY  - 2023
DA  - 2023/11/27
TI  - Semantic Segmentation of Rice Disease Images based on DeepLabV3+
BT  - Proceedings of the 2023 International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2023)
PB  - Atlantis Press
SP  - 935
EP  - 947
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-300-9_95
DO  - 10.2991/978-94-6463-300-9_95
ID  - Li2023
ER  -