Change Detection for Under Vehicle Surveillance

Project: Change Detection in Vehicle Bottom Images

Goal:

Currently, the under vehicle surveillance system is not able to automatically detect the differences between the synthetic image of a vehicle bottom taken during the pass and the image stored as reference in the local PACS (picture archiving and communication system). In this project, I aim to develop a fully automated Change Detection Module that can identify the change on screen by highlighting the differences between the current vehicle pass imagery and its corresponding reference image.

Method:

Image Enhancement; Histogram Equalization; Image Filter and Resize; Large displacement non-rigid image registration; Shape Analysis

Programming with C++ and OpenCV 3.0. The UI is created with C#.

Examples:

Others:

This method can be used for other applications related with change detection.