Rotation Aware Object Detection Model with Applications to Weapons Spotting in Surveillance Videos

Detection of weapons automatically is very important for improving the security and prosperity of people, in any case, it is a troublesome undertaking due to huge assortment of size, shape and presence of weapons. View point varieties what's more, impediment likewise are the reasons which makes this errand more troublesome. Further, the present detection algorithms of objects process rectangular areas, anyway a thin and long rifle may truly cover simply a little part of zone and the rest may contain unessential subtleties. To beat this issue, we propose a deep learning based model for detection of weapons with orientation, which not only gives rotation aware bound box but also improves the detection performance. The proposed model provides orientation with the help of angle classification by dividing angle into eight different classes. To train our model for weapon recognition another new dataset containing of around 6400 pictures is assembled from the web and afterward manually annotated that. We also provide three standard horizontal annotation format of our dataset as ground truth along with oriented ground truth for further exploration in future. The proposed model is assessed on this dataset, also, the near investigation with off-the rack object indicators yields predominant execution of proposed model, estimated with the standard assessment procedures. The dataset and the model implementation are made publicly available at this link: https://bit.ly/2TyZICF.

I. Introduction

Weapon hostility occurrences are to a great extent spread across the globe, what’s more, are being seen at extending repeat [1] , [2] and [3] . Reliably, around a quarter-million people have lost their lives because of gun savagery [4] . Steps for gun control don’t appear to be compelling in spite of such an enormous number of deplorable occasions. Gun mercilessness reliably covers the globe, and it is making ominous consequences for humankind. The issue ought to be tended to logically for the progression of general prosperity, individual’s well-being, and security. Recently strong voices have been raised for legitimate data supported exploration to prevent weapon violence [5] and to finance research for such tasks. Over the world, security offices, government, and private institutions have extended the utilization of observation frameworks to ensure the lives of individuals, secure structures, and monitor the business zone. As in this globalized world, there are more individuals to individuals contacts yet at the equivalent time the weapon savagery because of mutual hatred is likewise normal. Thus, a proficient firearm detection technique attached in the surveillance system to turn away weapon violence or make a prompt and timely response is unavoidable [6] , [7] , [8] , [9] and [10] . Such advances would not just ensure security yet also help the economy because of low clinical expense in light of timely response because of the early discovery of weapons through the computerized automated mechanism.