Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM69122
Volume:
10 |
11 |
Nov 2026
Pages:
1-7
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Computer Science and Engineering
Visvesvaraya Technological University, Kalaburagi, India
Computer Science and Engineering
Visvesvaraya Technological University, Kalaburagi, India
Fire accidents in residential environments pose a serious threat to human life and property. Early detection of fire and smoke is essential to prevent disasters and minimize damage. Traditional fire detection systems rely primarily on sensors such as heat detectors and smoke sensors, which often respond only after the fire has already intensified. These systems are also prone to false alarms caused by environmental conditions such as steam, dust, or cooking smoke. In recent years, advances in artificial intelligence and computer vision have made it possible to detect fire hazards by analyzing visual data from images and surveillance cameras. This paper presents a deep learning–based system for early fire and smoke detection in homes. The proposed system uses a Convolutional Neural Network (CNN) to classify images into three categories: fire, smoke, and normal conditions. The dataset used for training contains images representing these three classes, enabling the model to learn distinctive visual patterns such as flame color, smoke diffusion, and environmental texture. Image preprocessing techniques including resizing and normalization are applied to improve model accuracy. A web-based interface developed using the Flask framework allows users to upload images and obtain detection results instantly. The system provides prediction labels, confidence scores, and performance evaluation metrics such as accuracy, precision, recall, and F1-score. A confusion matrix is also generated to analyze model performance. Experimental results demonstrate that the proposed system can accurately identify fire and smoke conditions while reducing false alarms. The system provides a cost-effective and scalable solution for improving residential fire safety and can be extended to real-time video monitoring systems in future work.
KEYWORDS
This research presents a deep learning–based system for early fire and smoke detection in residential environments. By combining convolutional neural networks with a web-based interface, the system provides an effective solution for analyzing images and identifying potential fire hazards. Experimental results confirm that the proposed model achieves reliable performance in detecting fire and smoke conditions. The system reduces false alarms and offers an accessible approach for safety monitoring without requiring specialized hardware. Future improvements may include integrating real-time video analysis, mobile notifications, and IoT-based alarm systems to further enhance the system’s capabilities. The proposed framework demonstrates the potential of artificial intelligence in improving safety and preventing fire-related disasters.
[1] Dey, N., Ashour, A. S., Shi, F., Sherratt, R. S., & Gospodinova, E., “Developing residential fire detection systems using machine learning algorithms,” International Conference on Inventive Computing and Informatics, pp. 440–445, 2017. [2] Wu, X., Huang, Y., & Chen, L., “Deep learning-based fire detection using convolutional neural networks,” Computers in Biology and Medicine, vol. 103, pp. 220–227, 2018. [3] Mohan, S., Thirumalai, C., & Srivastava, G., “Effective heart disease prediction using hybrid machine learning techniques,” IEEE Access, vol. 7, pp. 81542–81554, 2019. [4] Jin, C., Wang, T., Zhao, S., Liu, H., & Zhang, J., “Video fire detection methods based on deep learning: A review,” IEEE Access, vol. 8, pp. 124–135, 2020. [5] Zhang, M., “AI-based fire and smoke detection using multimedia information analysis,” IEEE Transactions on Multimedia, vol. 22, no. 10, pp. 2624–2634, 2020. [6] Ramani, N., & Gopinath, S., “Forest fire detection using YOLO-based deep learning framework,” International Journal of Computer Applications, vol. 177, no. 43, pp. 12–18, 2020. [7] Saponara, S., Elhanashi, A., & Gagliardi, A., “Real-time video fire and smoke detection using CNN in surveillance systems,” IEEE Sensors Journal, vol. 21, no. 12, pp. 13402–13410, 2021. [8] García, S. A., Molina, J., Benítez, A., & Molina, I., “Deep learning-based early forest fire detection using satellite imagery,” Remote Sensing, vol. 13, no. 6, pp. 1–15, 2021. [9] Yang, M., Qian, S., & Wu, X., “Real-time fire and smoke detection using cloud-edge collaborative architecture,” IEEE Internet of Things Journal, vol. 9, no. 4, pp. 2785–2796, 2022. [10] Shukla, D., & Yadav, R. K., “Deep learning-based fire detection system for visually impaired individuals,” International Journal of Advanced Computer Science and Applications, vol. 14, no. 3, pp. 123–129, 2023.
Publication Date
11/19/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
3
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...