Deep Learning-Based Image Classification for Major Mosquito Species Inhabiting Korea

• Published in: Insects (Basel, Switzerland) Vol. 14; no. 6; p. 526
• Main Authors: Lee, Sangjun, Kim, Hangi, Cho, Byoung-Kwan
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 05.06.2023; MDPI
• Subjects: Accuracy; artificial intelligence; Artificial neural networks; Biological monitoring; Biological research; Biology, Experimental; Cameras; Classification; Color; Counting methods; Culicidae; Deep learning; Dengue fever; Embedded systems; Encephalitis; Error analysis; Fluorescence; Human error; Identification and classification; Identification methods; Image acquisition; Image analysis; Image classification; Image enhancement; image identification; Image processing; Imaging systems; Imaging systems in biology; Infections; Infectious diseases; Insects; Labor; Machine learning; Malaria; Methods; Monitoring; mosquito; Mosquitoes; Neural networks; Object recognition; Pest control; Populations; Preempting; Species; Tropical diseases; Vector-borne diseases; Vectors (Biology); West Nile virus; Zika virus; South Korea; United States > US; deep learning; image identification; artificial intelligence; mosquito
• ISSN: 2075-4450; 2075-4450
• DOI: 10.3390/insects14060526

Mosquitoes are one of the deadliest insects, causing harm to humans worldwide. Preemptive prevention and forecasting are important to prevent mosquito-borne diseases. However, current mosquito identification is mostly conducted manually, which consumes time, wastes labor, and causes human error. In this study, we developed an automatic image analysis method to identify mosquito species using a deep learning-based object detection technique. Color and fluorescence images of live mosquitoes were acquired using a mosquito capture device and were used to develop a deep learning-based object detection model. Among the deep learning-based object identification models, the combination of a swine transformer and a faster region-convolutional neural network model demonstrated the best performance, with a 91.7% F1-score. This indicates that the proposed automatic identification method can be rapidly applied for efficient analysis of species and populations of vector-borne mosquitoes with reduced labor in the field.