This is the project in SCUT Digitial Image Processing (2023 Fall). We used several difference method to detect motion. This project provides well-designed GUI.
Just make sure you have all the requirements needed. Run MotionDetector.py, then you will see the GUI widget. Firstly choose the folder that all your testing images are in. Then, you can change several parameters to see the difference of each method. The stable results start from the 6th image in your folder, because the Background-Modeling Method needs to be trained for several images.
You can download the datasets from Scene Background Initialization (SBI) dataset. You may want to select Highway II in SBI dataset for testing, because it performs better.