Early Fire Detection and Alert System using Modified Inception-v3 under Deep Learning Framework

Biswas, Anuraag; Ghosh, Swarup Kr; Ghosh, Anupam

doi:10.1016/j.procs.2023.01.200

Cited by 9 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Te temporal and spatial dynamic fre textures were analyzed in [20] using 2D and 3D wavelet fragmentation. In addition, the authors in [21][22][23][24][25][26] discussed machine learning and deep learning methods for detecting forest fres. Tese studies highlight the diverse range of techniques and algorithms employed in fre detection research, showcasing advancements in accuracy, efciency, and real-time performance.…”

Section: Integration With Iot and Smartmentioning

confidence: 99%

Multisensor Fuzzy Logic Approach for Enhanced Fire Detection in Smart Cities

Vaegae,

Annepu,

Bagadi

et al. 2024

Journal of Optimization

View full text Add to dashboard Cite

As smart cities expand rapidly, the demand for strong fire protection has become even more essential. A significant challenge lies in ensuring that fire detection systems integrate seamlessly into our modern infrastructures. Leveraging multisensor systems can yield reliable data on potential fires. Particularly in smart buildings, the effectiveness of multisensor fire detection algorithms becomes paramount. This research introduces a fuzzy logic-driven method that harnesses the power of smoke, flame, and temperature sensors. While smoke, flames, and elevated temperatures are primary fire indicators, they can manifest concurrently or sequentially. We have amalgamated various fuzzy logic rules (IF-THEN structures) to gauge the intensity of fires. When a fire ignites, the sensors spring into action, identifying its source and promptly notifying users via the Internet and GSM modems. Moreover, they relay the fire’s precise geographic coordinates to fire departments. The fire’s status will be consistently updated on a dedicated online portal. It was observed from our results that the system proficiently sends fire alerts to residents, and the fire status undergoes regular updates at 45-second intervals. This refreshing is initiated by the identification of a designated percentage of smoke and flame. These outcomes validate the system’s effectiveness in enhancing the precision and responsiveness of fire detection capabilities.

show abstract

Section: Integration With Iot and Smartmentioning

confidence: 99%

Multisensor Fuzzy Logic Approach for Enhanced Fire Detection in Smart Cities

Vaegae,

Annepu,

Bagadi

et al. 2024

Journal of Optimization

View full text Add to dashboard Cite

show abstract

“…On an extensive and varied freshly constructed dataset, the suggested method correctly predicted the Smoking and Non-Smoking images with an accuracy of 96.87%, 97.32% precision, and 96.46% recall. To detect fire, an improved Inception-V3 has been proposed [23] on the fire and smoke images dataset. This model includes a new optimizing function that effectively lowers the cost of computation.…”

Section: Related Workmentioning

confidence: 99%

An improved fire detection approach based on YOLO-v8 for smart cities

Talaat

ZainEldin

2023

Neural Comput & Applic

185

View full text Add to dashboard Cite

Fires in smart cities can have devastating consequences, causing damage to property, and endangering the lives of citizens. Traditional fire detection methods have limitations in terms of accuracy and speed, making it challenging to detect fires in real time. This paper proposes an improved fire detection approach for smart cities based on the YOLOv8 algorithm, called the smart fire detection system (SFDS), which leverages the strengths of deep learning to detect fire-specific features in real time. The SFDS approach has the potential to improve the accuracy of fire detection, reduce false alarms, and be cost-effective compared to traditional fire detection methods. It can also be extended to detect other objects of interest in smart cities, such as gas leaks or flooding. The proposed framework for a smart city consists of four primary layers: (i) Application layer, (ii) Fog layer, (iii) Cloud layer, and (iv) IoT layer. The proposed algorithm utilizes Fog and Cloud computing, along with the IoT layer, to collect and process data in real time, enabling faster response times and reducing the risk of damage to property and human life. The SFDS achieved state-of-the-art performance in terms of both precision and recall, with a high precision rate of 97.1% for all classes. The proposed approach has several potential applications, including fire safety management in public areas, forest fire monitoring, and intelligent security systems.

show abstract

“…This article builds a deep-learning image-based fire and smoke detector. Inception-V3 [11] is modified for smoke-filled fire images. The new optimization function reduces computation costs.…”

Section: Related Workmentioning

confidence: 99%

Revolutionizing Automotive Parts Classification Using InceptionV3 Transfer Learning

Hindarto

2023

ijsecs

View full text Add to dashboard Cite

This study presents a novel methodology for classifying automotive parts by implementing the Transfer Learning technique, utilizing the InceptionV3 architecture. We use a proprietary dataset encompassing diverse categories of automotive components for training and evaluating the model. The experimental findings demonstrate that this approach attains a performance accuracy level of 93.78% and a loss rate of 0.2938. The efficacy of InceptionV3 Transfer Learning in addressing the intricacies associated with automotive parts classification is demonstrated through its utilization of pre-existing knowledge from diverse domains. The resultant model reflects a capacity to accurately discern spare parts, thereby enhancing the efficiency of the automotive inventory management process. Utilizing InceptionV3 Transfer Learning in this scenario yields a notable and favorable outcome, thereby revolutionizing the traditional framework of automotive parts categorization. The model's efficacy in enhancing the efficiency and accuracy of automotive inventory management is evidenced by its achievement of a notable precision level and a minimal loss rate. The implications of these findings are significant in addressing intricate classification challenges within the automotive industry. They pave the way for utilizing intelligent technologies to optimize parts identification and management processes. This study establishes a foundation for a novel approach to comprehending and applying categorization systems for automotive components. This is achieved by harnessing the capabilities of Transfer Learning using the InceptionV3 model.

show abstract

Early Fire Detection and Alert System using Modified Inception-v3 under Deep Learning Framework

Cited by 9 publications

References 12 publications

Multisensor Fuzzy Logic Approach for Enhanced Fire Detection in Smart Cities

Multisensor Fuzzy Logic Approach for Enhanced Fire Detection in Smart Cities

An improved fire detection approach based on YOLO-v8 for smart cities

Revolutionizing Automotive Parts Classification Using InceptionV3 Transfer Learning

Contact Info

Product

Resources

About