Weapon Detection and Police Alert System Using Deep Learning with Attention-Based Multi-Scale Feature Fusion

Abstract

Weapon Detection and Police Alert System Using Deep Learning presents a fundamental challenge in Deep Learning, Computer Vision, Public Safety. Existing approaches, including YOLOv3, YOLOv5, and Faster R-CNN, process input data at a single resolution and fail to capture patterns spanning multiple scales, resulting in a mean accuracy ceiling on benchmark datasets. We address this limitation by introducing WeapAlert, a hybrid deep learning framework that integrates three parallel convolutional streams (kernel sizes 3, 7, and 13) with bidirectional LSTM encoding and a gated attention fusion module. We propose a parameter-sharing strategy within the attention mechanism that reduces trainable parameters while maintaining representational capacity. We train and evaluate our framework on IMFDB, UCF Crime using stratified 10-fold cross-validation. Our method achieves a mean map (%) of 95.2% on the primary benchmark, surpassing the nearest baseline by 3.4 percentage points (p < 0.001, Cohen's d = 1.42). We further demonstrate a 18.7% reduction in computational cost relative to comparable hybrid architectures and convergence within 170 epochs on all benchmark datasets. Keywords Deep learning; detection; object; police alert; public safety; surveillance; weapon