Application Of ADNN For Background Subtraction In Smart Surveillance System

TL;DR

This paper presents an intelligent surveillance system that leverages ADNN for efficient background subtraction, motion detection, and anomaly detection, reducing data labeling needs and improving performance on unseen videos.

Contribution

It extends the ADNN approach to a comprehensive surveillance system with motion and anomaly detection capabilities, enhancing robustness and efficiency.

Findings

Effective background subtraction using ADNN

Improved motion detection accuracy

Successful anomaly detection on trimmed videos

Abstract

Object movement identification is one of the most researched problems in the field of computer vision. In this task, we try to classify a pixel as foreground or background. Even though numerous traditional machine learning and deep learning methods already exist for this problem, the two major issues with most of them are the need for large amounts of ground truth data and their inferior performance on unseen videos. Since every pixel of every frame has to be labeled, acquiring large amounts of data for these techniques gets rather expensive. Recently, Zhao et al. [1] proposed one of a kind Arithmetic Distribution Neural Network (ADNN) for universal background subtraction which utilizes probability information from the histogram of temporal pixels and achieves promising results. Building onto this work, we developed an intelligent video surveillance system that uses ADNN architecture for motion detection, trims the video with parts only containing motion, and performs anomaly detection on the trimmed video.