My Research

Research: Automatic Traffic Red-
Light Violation Detection Using AI

I’ve had a chance to collaborate with Ph.D Le Quang Thao at the University of Technology and Science. Our research is the design of a traffic signal violation detection system using machine learning that learns to prevent the increasing number of road accidents. The system is optimized in terms of accuracy by using the region of interest and location of the vehicle with a red-signal state. By modifying some parameters in the YOLO 5s and retraining the COCO dataset, we can create a model which can be predicted with a high accuracy of 82% for vehicle identification, 90% for traffic signal status change and up to 86% for violation detection. This can be used for red light violation detection which will help the traffic police on traffic management.

published LINK

Research: Pest Early Detection in
Greenhouse Using Machine Learning

Continuing to work with Ph.D Le Quang Thao, we have completed the research about Pest Early Detection in Greenhouses. Greenhouses are considered to be a favorable artificial environment separated from the outside. However, pests can still exist by the same plant sources that bring the pathogen. The conditions and abundant food in a greenhouse provide a stable environment for the pest development. Normally, the natural enemies that serve to keep pests under control outside are not present in the greenhouse, pest situations often develop in this indoor environment more rapidly and with greater severity than outdoors. Early detection and diagnosis of pests and diseases are key to managing greenhouse pests as well as selecting and applying appropriate pesticides when needed. The aim of this invention is to develop an intelligent pest early detection system using a convolutional neural network in the greenhouse. By using a pre-trained disease recognition model, we were able to perform deep transfer learning to produce a network that can predict with the precision above 90%.

PRODUCT LINK