Moving Object Detection Using Background Subtraction Algorithms

Table of Contents

1. INTRODUCTION

1.1 OBJECTIVE

1.2 APPLICATIONS

1.3 LITERATURE SURVEY

1.4 ORGANIZATION OF THE REPORT

2. BLOCK DIAGRAM AND CHALLANGES

2.1 GENERAL STEPS FOR OBJECT DETECTION

2.2 CHALLENGES

3. STUDY OF DIFFERENT BACKGROUND SUBTRACTION ALGORITHMS

3.1 SIMPLE BACKGROUND SUBTRACTION METHOD

3.2 MEAN FILTERING METHOD

3.3 MEDIAN FILTERING METHOD

3.4 W4 SYSTEM METHOD

3.5 FRAME DIFFERENCING METHOD

3.6 RUNNING GAUSSIAN AVERAGE MODEL

3.7 GAUSSIAN MIXTURE MODEL

3.8 EIGENBACKGROUND

4. COMPARISION OF BACKGROUND SUBTRACTION ALGORITHMS

5. OPTICAL FLOW

5.1 THE SMOOTHNESS CONSTRAINT

5.2 DETERMINING OPTICAL FLOW USING HORN - SCHUNCK

5.3 ESTIMATION OF CLASSICAL PARTIAL DERIVATIVES

5.4 EXPERIMENT RESULTS

6. COMBINE GMM & OPTICAL FLOW

7. SHADOW DETECTION

7.1 HSV/HSI MODE

7.2 SHADOW DETECTION

8. CONCLUSION AND FUTURE WORK

Research Objectives and Core Themes

The primary objective of this dissertation is to present an operational computer video system designed for real-time moving object detection and tracking. The research focuses on evaluating, implementing, and improving various background subtraction algorithms to address challenges such as illumination changes, noise, and the requirement for high-accuracy tracking in surveillance applications.

• Systematic review and performance evaluation of common background subtraction algorithms.
• Implementation of motion detection techniques on the MATLAB software platform.
• Comparative analysis of algorithms based on speed, memory usage, and detection accuracy.
• Development of a combined GMM and Optical Flow approach to improve detection robustness against lighting variations.
• Application of HSV color space modeling for effective shadow detection and removal.

Book Excerpt

Summary of Chapters

1. INTRODUCTION: This chapter introduces the necessity of motion detection in surveillance and outlines the organization of the research report.

2. BLOCK DIAGRAM AND CHALLANGES: This chapter details the general workflow of object detection systems and identifies critical challenges such as illumination changes and moving shadows.

3. STUDY OF DIFFERENT BACKGROUND SUBTRACTION ALGORITHMS: This chapter discusses various segmentation methods, ranging from Simple Background Subtraction to more complex techniques like GMM and Eigenbackground.

4. COMPARISION OF BACKGROUND SUBTRACTION ALGORITHMS: This chapter provides a quantitative comparative analysis of the studied algorithms in terms of computational speed and memory requirements.

5. OPTICAL FLOW: This chapter explains the theoretical basis of optical flow methods and their application in motion estimation using the Horn-Schunck algorithm.

6. COMBINE GMM & OPTICAL FLOW: This chapter introduces a hybrid approach that integrates GMM and Optical Flow to mitigate the negative effects of sudden illumination changes.

7. SHADOW DETECTION: This chapter explores the use of HSV color space to identify and suppress shadows that otherwise cause false detections.

8. CONCLUSION AND FUTURE WORK: This chapter summarizes the experimental findings and suggests future improvements, such as the use of pyramid approaches and handling moving camera scenarios.

Keywords

Background Subtraction, Moving Object Detection, Optical Flow, Gaussian Mixture Model, Surveillance Systems, Image Segmentation, Shadow Detection, MATLAB, Horn-Schunck Algorithm, Principal Component Analysis, Eigenbackground, Motion Tracking, Computer Vision, Thresholding, Illumination Changes

Frequently Asked Questions

What is the primary focus of this dissertation?

The dissertation focuses on the development and evaluation of computer vision systems for real-time moving object detection and tracking using various background subtraction algorithms.

What are the core thematic areas covered in this report?

The core themes include algorithmic performance analysis, comparative studies on computational complexity, motion detection methods (GMM, Optical Flow), and techniques for shadow suppression in surveillance environments.

What is the main research goal?

The goal is to analyze existing moving object detection algorithms and improve upon their limitations, specifically regarding their sensitivity to noise and sudden lighting fluctuations.

Which scientific methods are primarily utilized?

The research utilizes statistical modeling, background subtraction techniques, optical flow estimation based on the Horn-Schunck algorithm, and color-space analysis (HSV) for image processing.

What does the main body of the work cover?

The main body examines several algorithms (Simple BS, Mean/Median Filtering, W4, Frame Differencing, GMM, Eigenbackground), compares their performance metrics, and proposes a combined GMM-Optical Flow method.

Which keywords characterize this research?

The work is characterized by terms such as Background Subtraction, GMM, Optical Flow, Surveillance, and Shadow Detection.

How does the proposed combined approach improve performance?

The combined approach merges the accurate segmentation capabilities of GMM with the motion-robustness of Optical Flow to minimize false detections caused by illumination changes.

What role does the HSV color space play in this study?

HSV modeling is used specifically for shadow detection, as it effectively separates color information from intensity, allowing for the isolation of shadows that are often misidentified as moving objects.