VigilantAI: Real-time detection of anomalous activity from a video stream using deep learning

In an era where artificial intelligence (AI) solutions are increasingly integrated into various sectors, we have utilized Artificial Intelligence for enhancing public safety through real-time detection of illegal activities such as robberies and threats at gunpoint using CCTV footage. With the proliferation of deep learning in object detection, the study focuses on deploying the YoloV5 model, trained on a custom dataset compiled from diverse CCTV sources and movies, to identify specific criminal actions. One of the major problems faced in this field is the availability of a large robust labeled dataset on which a deep learning model can be trained. For this purpose, we have created our own dataset by converting various CCTV footage and movies into images, and then labeling them with the correct class. In addition, we also augmented data by using various data augmentation techniques for the chosen images. This dataset, enriched through augmentation techniques and annotated with bounding boxes, allows for the precise detection of threats, achieving an accuracy rate of 85%. Our system stands out by not only spotting these activities but also by instantly alerting security personnel, facilitating a rapid response to potentially dangerous situations. This capability is important for law enforcement agencies worldwide, offering them an advanced tool to act swiftly and prevent crimes, thereby enhancing public security. The essence of our work demonstrates the practical application and significant impact of AI in bolstering security measures, providing a solid foundation for future enhancements in the field. Through this initiative, we aim to foster a safer environment in public spaces, reducing crime rates and increasing the general public’s sense of safety.

I. Introduction

Detecting illegal activities such as robbery and gunpoint events and conducting surveillance is a challenging task that often requires a significant workforce, leading to potential human errors. Machine learning and Artificial Intelligence [1] have been very popular in automation of several similar tasks in computer vision ranging from generative AI [2] to object detection [3]–[5] and image segmentation [6]–[8]. Machine learning has been especially useful for multiple practical applications [9], [10], for example, in crop disease detection [11], intrusion detection [12], cyber bullying [13], medical applications [14], [15], and linguistics [16], [17]. As shown in other domains, utilizing computer vision and artificial intelligence offers a more accurate alternative for illegal activity detection. This project focuses on using AI models to detect specific illegal activities, aiming to enhance surveillance and law enforcement. The Yolov5s [18] object detection model is employed, and trained on a custom-designed dataset for illegal activity detection. Python programming language, Google Colab, and PyCharm are utilized for model training and deployment. For seamless deployment on IoT devices, the PyThorce model is chosen due to its ease of use.