The design and development of an automatic detection system of the oil palm bagworms are based on deep convolutional neural networks (Conv. Net) and ordinary camera. It consists of motion-tracking and false color analysis to detect and count the number of living and dead larvae and pupae of M. plana population per frond, corresponding to three major groups or sizes classification, which is applied using ground basis application and close system.

Main Researcher: Mohd Najib Ahmad
Email: mnajib@mpob.gov.my

https://www.youtube.com/watch?v=zv5rPupYemc&feature=youtu.be