Detection of Aedes aegypti larvae using single shot multibox detector with transfer learning

The flavivirus epidemiology has reached an alarming rate which haunts the world population including Malaysia. In fact, World Health Organization has proposed and practised many methods of vector control through environmental management, chemical and biological orientations but still cannot fully overcome the problem. This paper proposed a detection of Aedes Aegypti larvae in water storage tank using Single Shot Multibox Detector with transfer learning. The objective of the study was to acquire the training and the performance metrics of the detection. The detection was done using SSD with Inception_V2 through transfer learning. The experimental results revealed that the probability detection scored more than 80% accuracies and there was no false alarm. These results demonstrate the effectiveness of the model approach.