Object detection and classification on X-ray baggage scanning dataset using neural network
- Author Harsh Shekhar
- DOI https://ww
- Country : India
- Subject : Computer Science and Engineering
Keywords : security, collaboration, special education, disability
Abstract
The aim for this project was to create an automatic system for security baggage scanning at the airports by training and testing three different neural network algorithms namely FRCNN, RFFCN, YOLOv2 and I also wanted to make the system faster for training purpose for which I made some changes on FRCNN code and also on YOLOv2 and then tested these algorithms on x-ray baggage image dataset which would include various firearm components such as guns, knives, pliers and wrenches, which would be detected by the algorithms and then these algorithms would be compared on various parameters. I have divided the comparing parameters as primary parameters and secondary parameters where primary parameters would be the one which would be manually analyzed from the resultant output testing images and the secondary parameters would be the ones which would be calculated from the primary parameters. These parameters were then compared for each of the firearm component separately to conclude the best algorithms out of the three for x-ray image dataset scanning.
Comments
No have any comment !