Mosquito Breeding Grounds Detection Using Deep Learning Techniques

Mosquito-borne diseases, including malaria, dengue, chickenpox, yellow fever, and jaundice, are on the rise daily. To combat this, it is crucial to identify and eliminate mosquito breeding sites. This paper presents a novel approach to mosquito breed ground detection using deep learning models with different architectures, including YOLO, Mask R-CNN, and Transformers. We compare the performance of these models on a Mosquito Breed Ground (MBG) dataset using various optimizers and evaluate their Mean Average Precision (MAP) as a measure of accuracy. Our results show that the deep learning models outperform traditional methods for mosquito breed ground detection and achieve a high MAP of 0.92. Furthermore, a comprehensive analysis of the performance of each model is provided.