A joint research team at CNU built an image dataset of 41 representative exotic weed species existing in Korea.

Since 2018, CNU's Intelligent Electronics Lab (Director: Professor Kim Jin-young) has collaborated with Professor Lee Yong-ho (Hankyung University), Professor Kim Jung-gu (Chungnam National University), Professor Kim Seung-cheol (Sungkyunkwan University), Professor Jeong Young-jae (Shingyeong University), and the Dr. Kim Chang-seok (National Institute of Food Science and Technology). The joint research team has built an image dataset and developed a deep learning model capable of identifying plants and weeds based on images, including the open Plant Seedlings Dataset at Aarhus University in Denmark.

This research team published a paper in ELSEVIER Computers and Electronics in Agriculture, which is among the SCIE's top 10% of journals.

In addition, the joint research team developed a deep-learning training model to effectively solve the imbalance problem of the weed dataset collected disproportionately for each recognized species, and the research results were published in MDPI Applied Sciences.

The research team said, "Based on many years of research, we accumulated relevant data to develop and test an Android application that recognizes non-native weeds for efficient management of agricultural land this year."

Currently, the research team is carrying out the project of Developing non-native weeds image identification system, with cooperation from the Rural Development Administration, the National Institute of Food Science and Technology, and the Future Environmental Ecology Research Institute Co., Ltd.