ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Conference Paper Image-Based Predictive Model to Optimize Drying Endpoints in the Chili Pepper Drying Process
Cited 0 time in scopus Download 63 time Share share facebook twitter linkedin kakaostory
Authors
Dasong Yu, Aekyeung Moon
Issue Date
2024-09
Citation
Asia Service Sciences and Software Engineering Conference (ASSE) 2024, pp.1-8
Language
English
Type
Conference Paper
DOI
https://dx.doi.org/10.1145/3702138.3702154
Abstract
Traditional agricultural drying methods have been a fundamental part of crop post-processing, yet they are frequently characterized by their labor-intensive nature and susceptibility to significant crop losses due to inconsistent drying practices. This paper introduces image-based drying crop predictive models designed to automate and optimize the chili pepper drying process by continuously monitoring the crops’ drying state. The proposed model leverages crop images, weight measurements, and time data to accurately assess the drying state of chili peppers. By systematically analyzing visual characteristics and temporal weight changes, the model learns to identify key indicators of the drying process. The performance of the model, based on the ResNet architecture, achieved an MAE of 0.025, RMSE of 0.032, and R2 score of 0.985. These results indicate high precision in predicting the drying state, which can significantly enhance the reliability and efficiency of the drying process. These technologies provide continuous automatic monitoring and enable the analysis of crop drying conditions using collected image datasets and real-time adjustments, thereby enhancing the efficiency of traditional, experience-based, laborious agricultural drying operations.
KSP Keywords
Chili pepper, Continuously monitoring, Drying condition, Drying methods, Drying process, Image-based, Key indicators, Post-Processing, Predictive model, Proposed model, Real-time
This work is distributed under the term of Creative Commons License (CCL)
(CC BY)
CC BY