An Automatic Detection of Military Weapons Using multi-Level CNN Architecture

This study elucidates an automatic detection of weapons used in military based on deep learning. It utilizes multi-level CNN architecture with three layers of CNN to classify three kinds of visual images for detection of weapons in an image. This proposed system achieved an accuracy of 97.7
^%
for the image classification. This experiment was performed on datasets of 2700 images using Tensorflow and Keras.