Chonnam National University Joint Research Team Develops Invasive Weed Identification System

The joint research team at Chonnam National University is presenting the development of an invasive weed image identification app and system. Photo by Chonnam National University

[Asia Economy Honam Reporting Headquarters Reporter Cho Hyung-joo] A joint research team at Chonnam National University has developed an application and system that can identify invasive weeds through images.

According to Chonnam National University on the 1st, the joint research team from the Intelligent Electronics Laboratory (Professor Kim Jin-young) recently revealed this at the 2021 Inter-Ministerial Performance Presentation of the Invasive Species Management Technology Development Project, hosted by the Ministry of Environment and organized by the Korea Environmental Industry & Technology Institute (KEITI), Rural Development Administration, Quarantine Headquarters, Korea Forest Service, and Korea Forestry Promotion Institute.

The joint research team announced that based on four years of research from 2018 to 2021 on developing a deep learning-based invasive weed identification system, they built a database, developed an annotation tool for constructing a deep learning training database, and performed a deep learning-based algorithm for invasive weed image identification, ultimately developing an Android application and system for invasive weed image identification.

This system allows users to take photos of weeds in the field using their smartphone camera, edit the desired part, and then click the "Invasive Weed Identification" button, where the deep learning-based invasive weed image identification model analyzes and displays the results.

The server categorizes the input invasive weed images according to the identification results and also stores metadata such as the shooting date, time, and location information provided by the smartphone, thereby building a database for future invasive weed identification.

The research team consists of Dr. Kim Chang-seok from the National Institute of Crop Science, Professor Lee Yong-ho from Hankyong National University, Professor Kim Jung-gu from Chungnam National University, Professor Kim Seung-chul from Sungkyunkwan University, Professor Jung Young-jae from Shingyeong University, and Future Environment Ecology Research Institute Co., Ltd. (CEO Oh Young-joo). Since 2018, they have been developing the "Deep Learning-Based Invasive Weed Identification System."

This content was produced with the assistance of AI translation services.