International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 02 | Feb 2025
p-ISSN: 2395-0072
www.irjet.net
A CNN-Based Model for Detection of Onset of Indoor Fires Nwabueze, C. E1, Nwokorie, E. C2, Onyemauche, U. C3, Ayogu, I. I4 1, 2, 3,4 Department of Computer Science, Federal University of Technology Owerri, Nigeria.
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - This research focuses on creating and assessing
emergency response teams. This critical function is effectively carried out by fire monitoring and detection systems. Depending on factors such as the expected fire scenario, building type, occupancy, and fire severity, these systems are tailored to provide either general or specialized functionalities [3]. Fire monitoring involves observing and managing fire risks in a specific location based on parameters such as weather conditions, the state of flammable materials, and potential ignition sources. This process aids in implementing preventive measures and minimizing damage before a fire occurs. It also incorporates mechanisms to predict factors that may trigger a fire outbreak [5]. Fire detection, on the other hand, focuses on identifying fires in their early stages, allowing for the safe evacuation of occupants. When technology is applied to fire detection, implementing a robust and efficient algorithm becomes essential, as it plays a critical role in the overall firefighting process [6]. This study introduces a system designed for fire monitoring and detection using computer vision and convolutional neural networks (CNN). The aim is to enhance conventional fire detection systems, which typically rely on sensors for detecting smoke, temperature changes, and other fire byproducts. Traditional sensor-based systems, while useful, are often constrained by space and location, limiting their efficiency in certain environments [4]. In this research, a cost-effective computer vision approach is employed, utilizing high-definition (HD) cameras to capture real-time video images within the deployment area, enabling the prediction of potential fire situations. A CNN model was trained on a comprehensive dataset of indoor and outdoor fire images sourced from Kaggle, ensuring diverse representation to support robust generalization. The trained model's performance was rigorously evaluated using an independent dataset, with key metrics such as accuracy, precision, recall, and F1-score assessed to validate its effectiveness. Furthermore, the final prototype of the fire detection system was deployed as a user-friendly web-based platform. This platform, developed using Python, HTML, CSS, and the Flask framework, integrates real-time detection capabilities to provide an efficient and practical solution for fire monitoring and detection.
an artificial intelligence-driven model designed to detect fire incidents. The model was developed to identify fire occurrences in indoor settings using computer vision methodologies and convolutional neural networks (CNN). The CRISP-DM (Cross-Industry Standard Process for Data Mining) framework was applied throughout the project to guide the development and data analytics process. A dataset comprising 5,030 fire and non-fire images was sourced from Kaggle.com (2023). These 3- channeled images, with a resolution of 224 × 224 pixels, were processed using Python libraries such as TensorFlow, Keras, and NumPy. The model was trained and tested on a binary classification task to distinguish between fire and non-fire scenarios. Performance evaluation revealed a high accuracy rate of 98.77%. Detailed metrics included precision, recall, and F1-scores for both classes: non-fire instances achieved precision, recall, and F1-scores of 98.70, 99.12, and 98.70, respectively, while fire instances recorded scores of 98.18, 96.43, and 97.40, respectively. The confusion matrix provided further insights, with the fire class test set yielding true positives (TP) of 54, false positives (FP) of 1, false negatives (FN) of 2, and true negatives (TN) of 187. Similarly, the non-fire class test set recorded TP as 270, FP as 3, FN as 5, and TN as 2. These findings demonstrate the model's high efficacy in fire incident detection, while also identifying potential areas for enhancement. By integrating artificial intelligence with computer vision, the proposed model has the potential to significantly improve fire safety measures and enable rapid emergency responses.
Key Words:
Fire, Fire Detection, Fire Monitoring, Convolutional Neural Network, Computer Vision
1. INTRODUCTION Throughout history, fire incidents have posed significant risks to domestic and industrial properties, buildings, sites, and offices [1]. The consequences of fire outbreaks are often catastrophic, leading to substantial loss of lives and property. Beyond these immediate damages, fire incidents have inflicted financial burdens on nations, costing billions of dollars in repairs and reconstruction efforts [2]. To address these challenges, researchers and fire management agencies have continuously developed various methods for fire monitoring, detection, and response, utilizing advanced technologies and tools. According to the Northeast Document Conservation Centre (NEDCC), a key strategy for fire protection lies in identifying potential fire emergencies early and promptly alerting building occupants and
© 2025, IRJET
|
Impact Factor value: 8.315
2. RELATED WORKS To date, there is no universally optimized algorithm for fire detection using computer vision; instead, various studies have explored different approaches to model the fire detection problem, aiming to enhance accuracy and minimize false alarms. Existing research reveals that numerous algorithms have been developed and proposed for
|
ISO 9001:2008 Certified Journal
|
Page 110