Species identification has been performed using morphological features and molecular techniques between each species. Recently, advances in machine learning technology have been applied to animal and plant identification based on deep learning approach with images. In this study, we constructed the deep learning model to predict 16 species of marine ragworms using the Teachable Machine which is a web-based tool. We have trained the deep learning model with 865 images including key characters of 16 species and measured prediction accuracy using 3-fold cross validation. The results showed 94% sensitivity, 99% specificity, and 99% accuracy. The deep learning model for marine ragworms is freely available at forensicdna.kr/ML/ nereididae.
목차
Abstract Ⅰ. Introduction Ⅱ. Materials and Methods 1. Images collection and DLM construction 2. CNN architecture of MobileNet 3. Performance estimation of DLM Ⅲ. Results 1. Data cleaning and image correction 2. Construction of DLM 3. Performance estimation of model using 3-fold cross validation Ⅳ. Discussion Ⅴ. Acknowledgements Ⅵ. References
법과학 분야는 사회정의 구현에 있어 크나큰 가치가 있음에도 불구하고 우리나라에서는 이 분야에 대한 인식이 미흡하여 선진 외국에 비해 침체되어 있는 실정이다. 이에 우리나라에서도 법과학 분야와 관련 있는 학계, 연구기관, 수사기관 등 유관 단체들로 구성된 한국 법과학회를 창립하여 이 분야를 활성화 시켜 과학수사를 한층 더 발전시키기 위함을 목적으로 한다.