- Giáo viên hướng dẫn: thầy Huỳnh Tuấn Anh
- Sinh viên thực hiện: Lê Thị Phương Ngân - MSSV: 16520792.
Trường Đại học công nghệ thông tin, ĐHQG-TPHCM.
- Công nghệ: OpenCV 3.4, SVM (Support Vector Machine).
- Ngôn ngữ: C++, Python.
- Video Demo: tại đây
- Báo cáo: Fire Detection Report
- Window, Visual Studio 2019.
- OpenCV 3.4.1.
Mở Windows PowerShell: chạy lệnh:
cd Fire-Detection\Motion2Cpp\x64\Release\
./Motion2Cpp.exe [Video Path]
ví dụ:
./Motion2Cpp.exe \Fire-Detection\videoTest\test.mp4
I. Image Processing.
-
Background Subtraction.
-
Fire Color Filtering.
-
Mophological Processing.
-
Connected Component Labeling.
II. Foreground region analysis.
III. Fire Dynamic Behavior Analysis
IV. Fire Flow Energy Analysis
[1] G. Healey, D. Slater, T. Lin, B. Drda, and D. Goedeke. “A system for real-time fire detection,” IEEE International Conference on Computer Vision and Pattern Recognition, pp. 605 – 606, January 1993
[2] C. B. Liu and N. Ahuja ”Vision Based Fire Detection, IEEE International Conference on Pattern Recognition,Vol. 4, pp. 134 – 137, August 2004.
[3] H.D. Duong, D.D. Nguyen, L.T. Ngo, and D.T. Tinh, “On Approach to Vision Based Fire Detection Based on Type-2 Fuzzy Clustering,” IEEE International Conference on Soft Computing and Pattern Recognition, pp. 51-56, October 2011.