Deep Learning-Driven Low-Resolution Thermal Image Reconstruction for Enhanced CWSI Analysis

The timely detection of drought stress in chili plants is a prerequisite for the healthy life of the crop and maximum production. This study proposes a new method to address the problem with the help of deep learning for the purpose of effectively identifying drought resistance in chili plants and f...

Full description

Saved in:
Bibliographic Details
Main Authors: George, Anzen T, Deepak, Devithaa M, Kunnath, Gayatri Pradeep, Devi, Gowri Parvathy, S, Sarath
Format: Conference Proceeding
Language:English
Subjects:
Accuracy
CNN
Convolutional neural networks
Crops
CWSI
Droughts
ESPCN
Image resolution
Productivity
RDN
Resilience
SRCNN
Stress
Visualization
Water resources
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The timely detection of drought stress in chili plants is a prerequisite for the healthy life of the crop and maximum production. This study proposes a new method to address the problem with the help of deep learning for the purpose of effectively identifying drought resistance in chili plants and focus on the detail analysis of plant health. We intend to use deep learning frameworks to design Convolutional Neural Network (CNN) models to upscale the low-resolution thermal images of chili plants to their high-resolution images. CWSI is computed based on the generated images, further employed to evaluate drought stress in chili plants. A real-world dataset of chili plants under various water stress circumstances will be gathered in order to accomplish this. The Crop Water Stress Index(CWSI) can be computed using the high-resolution thermal images generated via the CNN models. Compared to conventional visual inspection techniques, CWSI offers a more objective and accurate assessment by providing a quantitative measure of a plant's water stress level. Our method seeks to greatly increase the accuracy of drought resistance evaluation in chili plants by enhancing the integrity and visual clarity of low-resolution thermal images. Improved accuracy and effectiveness in agricultural monitoring and resource management strategies will follow from this. Ultimately, it could empower farmers to make data-driven decisions about irrigation techniques, maximizing water use effectiveness and guaranteeing the productivity and long-term health of their chili crops.
ISSN:2996-5357
DOI:10.1109/ICESC60852.2024.10689781