Deteksi Api pada Video menggunakan Metode Multi-Feature Fusion dan CNN
Abstract
Kebakaran merupakan bencana yang memiliki dampak buruk bagi lingkungan dan juga menyebabkan kerugian yang signifikan terhadap harta benda serta kehidupan manusia. Deteksi api dengan perangkat sensor memiliki kelemahan di mana api harus berada dalam area tertentu untuk dapat memicu alarm. Pe- manfaatan computer vision untuk mendeteksi api dalam video menunjukan performa yang lebih baik dari pada deteksi menggunakan perangkat sensor. Deteksi api dengan computer vision memanfaatkan karak- teristik api seperti warna, gerakan dan perpindahan tempat. Karakter warna api menunjukan performa yang lebih baik dibandingkan dengan karakter lainnya, namun masih memiliki false positive yang tinggi sa- at ada objek yang memiliki warna menyerupai warna api. Pemanfaatan deep learning untuk menggantikan ekplorasi karakteristik api telah diteliti dan mampu menghasilkan prediksi yang lebih baik. Penelitian ini mencoba menggabungkan metode deteksi api dengan karakteristik gerak dan warna (multi-feature fu- sion) dengan metode deep learning CNN. Sistem yang dibangun memiliki akurasi 96,67% saat training dan mampu mengurangi deteksi false positive dari multifeature fusion.
Kata kunci—multi-feature fusion, CNN, deteksi api, computer vision, deteksi warna, deteksi gerak
References
P. Borges and E. Izquierdo. A Probabilistic Approach for Vision-Based Fire Detection in Videos. Circuits and Systems for Video Technology, IEEE Transactions on, 20:721–731, June 2010. doi: 10.1109/TCSVT. 2010.2045813.
Brushlinsky N.N, M. Ahrens, and S. Sokolov. World fire statistics. Technical Report 24, Center of Fire Statistics of CTIF, State Fire Academy of Emercom of Russia, Russia.
J. Fonollosa, A. Solo´rzano, and S. Marco. Chemical Sensor Systems and AssociatedAlgorithms for Fire Detection: A Review. Sensors, 18(2):553, Feb. 2018. ISSN 1424-
doi: 10.3390/s18020553. URL http://www.mdpi.com/1424-8220/18/2/553.
S. Frizzi, R. Kaabi, M. Bouchouicha, J.-M. Ginoux, E. Moreau, and F. Fnaiech.
Convolutional neural network for video fire and smoke detection. IECON 2016 - 42nd Annual Conference of the IEEE Industrial Electronics Society, pages 877–882, 2016.doi:10.1109/IECON.2016.7793196.
A. Gaur, A. Singh, A. Kumar, K. S.Kulkarni, S. Lala, K. Kapoor, V. Srivastava, A. Kumar, and S. C. Mukhopadhyay. Fire Sensing Technologies: A Review. IEEE Sensors Journal,
(9):3191–3202, May 2019. ISSN 1530-437X, 1558-1748, 2379-9153. doi: 10.1109/JSEN.2019.2894665. URL https://ieeexplore.ieee.org/document/8625538/.
F. Gong, C. Li, W. Gong, X. Li, X. Yuan, Y. Ma, and T. Song. A Real-Time Fire Detection Method from Video with Multifeature Fusion. Computational Intelligence and Neuroscience, 2019:1–17, July 2019.ISSN 1687-5265, 1687-5273. doi:10.1155/2019/1939171. URL
https://www.hindawi.com/journals/cin/2019/1939171/.
F. N. Iandola, S. Han, M. W. Moskewicz, K. Ashraf, W. J. Dally, and K. Keutzer. LEVEL ACCURACY WITH 50 X FEWER PARAMETERS AND < 0 . 5 MB MODEL SIZE. 2016.
B. Kim and J. Lee. A Video-Based Fire Detection Using Deep Learning Models. Applied Sciences, 9(14): 2862, July 2019.
ISSN 2076-3417. doi: 10.3390/app9142862. URL https://www.mdpi.com/2076-3417/ 9/14/2862.
S. Kru¨ger, M.-C. Despinasse, T. Raspe, K.No¨rthemann, and W. Moritz. Early fire detection: Are hydrogen sensors able to detect pyrolysis of house hold materials? Fire Safety Journal, 91:1059–1067, 2017.
K. Lee, Y.-S. Shim, Y. Song, S. Han, Y.-S. Lee, and C.-Y. Kang. Highly Sensitive
Sensors Based on Metal- Oxide Nanocolumns for Fire Detection. Sensors, 17(2):303, Feb.
ISSN 1424-8220. doi: 10.3390/ s17020303. URL http://www.mdpi.com/1424-8220/17/2/303.
K. Muhammad, J. Ahmad, Z. Lv, P.
Bellavista, P. Yang, and S. W. Baik. Efficient Deep CNN-Based Fire Detection
and Localization in Video Surveillance Applications. IEEE Transactions on Systems,
Man, and Cybernetics: Systems, 49(7):1419–1434, July 2019. ISSN 2168-2232. doi: 10.1109/TSMC.2018.2830099.
M. Senthil. Implications of Color Models in
Image Processing for Fire Detection.International Journal of Computer Applications, 179(35):38–41, Apr. 2018.
ISSN 09758887. doi:10.5120/ijca2018916801. URL
http://www.ijcaonline.org/archives/volume179/number35/vadivu-2018-ijca916801.pdf.
R. Sowah, A. R. Ofoli, S. Krakani, and S.Fiawoo. Hardware module design of a realtime multi-sensor fire detection and notification system using fuzzy logic. In
IEEE Industry Application Society Annual Meeting, pages 1–6, Vancouver, BC,
Canada, Oct. 2014. IEEE. ISBN 978-1-4799-
-8. doi: 10.1109/ IAS.2014.6978415.URL http://ieeexplore.ieee.org/document/6978415/.
X. Wang, Y. Li, and Z. Li. Research on flame detection algorithm based on multi -
feature fusion. In 2020 IEEE 4th Information Technology, Networking, Electronic and
Automation Control Conference (ITNEC), 2020. IEEE. ISBN 978-1-72814-390-3. doi:
1109/ITNEC48623. 2020.9084825. URLhttps://ieeexplore.ieee.org/document/9084825/.
Z. Xu and J. Xu. Automatic Fire Smoke Detection Based on Image Visual Features.
In Proceedings of the 2007 International Conference on Computational Intelligence and Security Workshops, CISW ’07, pages
–319, USA, 2007. IEEE Computer Society. ISBN 0-7695-3073-7.
N. Zaidi, N. Lokman, M. Daud, M.Achmad, and A. Khor. Fire recognition using RGB and YCbCr color space.
:9786–9790, Jan. 2015.
Q. Zhang, J. Xu, L. Xu, and H. Guo. Deep Convolutional Neural Networks for Forest Fire Detection. 2016. doi:10.2991/IFMEITA-16.2016.105.
