Automatic Threat Detection Using Deep Neural Networks
Saswati Soumya Tripathy, Laxmipriya Barik, Prajnya Paramita Sahu, Haraprasad Naik · 2021
Baggage inspection is a primary task for threat detection in any public facility area (such as Railway Station, Bus Terminal, and Airport). Most often these detections are done manually by security personnel that is prone to human error. In this paper, we have proposed a model of automatic threat detection mechanism that will produce a warning text message to the authority using X-ray images of baggage to prevent terrorist attacks. The proposed model is consisting of a Convolutional Neural Network with a Long Short-Term Memory (LSTM), which produces the real-time caption of the X-Ray Images. There will be an algorithm which will detect the keywords such as, Gun, Explosive, Knife, Blade, and Shuriken in the generated caption. Finally, a warning message can be generated to alert the security and authority to take precautionary measures. In this paper, we have used the GDX-ray dataset to train the model.