A Review on An Efficient Protocol to Capture UAV-image Frames and Detect the Fire and Smoke Using Image Features and Local Binary Pattern

Abstract

Lately fire outbreak is common issue happening in open space environments and the damage caused by these types of incidents is dangerous toward nature and human interest. Due to this the need of application for fire detection has increases in recent years. The accuracy of proposed algorithm is increased with decrease the non-smoke pixels of an image. Only true smoke pixels are analyzed and detected smoke on them. In this way the proposed algorithm is more accurate and computational time is decreased. The base algorithm’s accuracy is decreased due to video frames and that consumes so much time to detect true smoke pixels. So the comparison between the proposed algorithm and the base algorithm is more.

Introduction

Fire detection is essential for environmental and human safety. Video-based fire detection systems are increasingly preferred over traditional sensor-based systems due to their real-time performance, higher accuracy, and wide-area coverage, especially in open environments like forests or industrial zones.

Key Concepts:

1. Types of Detection Methods:

   • Smoke detection: Uses motion and color changes in video frames.

   • Flame detection: Uses visual (RGB), IR imagery, and motion analysis.

   • Both methods depend on ideal environmental conditions.

2. Challenges in Open Environments:

   • Smoke or flames may not be clearly visible due to distance or obstructions.

   • Smoke has no fixed color or shape, making detection harder.

3. Detection Criteria:

   • Systems analyze pixel color (RGB, YCbCr) and motion.

   • Smoke is identified through dispersion patterns and chrominance properties.

   • Early smoke is bluish-white; later smoke turns dark gray to black.

Literature Insights:

• Kim and Wang's approach: Three-step smoke detection accounting for camera movement, change detection, and ROI identification.

• Use of HSV color space, statistical modeling, Fourier transforms, and transparency features helps in reducing false positives.

• Final decision on fire presence is made by analyzing both smoke and flame characteristics.

Objectives:

• Improve early smoke detection through:

  1. Motion tracking of smoke using video.

  2. Boundary marking for accurate region detection.

  3. Reduction in false positives.

  4. Performance comparison with existing algorithms.

Methodology:

1. Color Detection: Classify smoke pixels using rule-based thresholds in RGB/YCbCr.

2. Area Detection: Track spread of smoke over time using frame analysis.

3. Motion Detection: Compare pixel changes between frames for movement.

4. Smoke Detection: Identify low-chrominance smoke with color evolution over time.

Proposed Algorithm Steps (MATLAB-based):

1. Import image from database.

2. Convert RGB to grayscale.

3. Analyze pixel-by-pixel for smoke color ranges.

4. Classify pixels as smoke or non-smoke.

5. Display the final detected output image.

Conclusion

The image frames to detect fire and smoke system is proposed in this work. The fire and smoke detection methods are applied on real time different database images. Initially, the true smoke pixels are detected to increase the accuracy of the algorithm and decrease the non-smoke pixels. Due to this the detection rate is increased and this system will be more accurate. The more images are taken from the database according to the requirements. It has been observed that the average retrieval efficiency is increased as feature set increases. Also, it has been observed that the true smoke detection rate is increased with increase in non-smoke pixels of an image. The result tables show the accuracy of the algorithm increases with decrease in false smoke detection. Furthermore, the main drawback of the proposed algorithm is that the completion time increases with decrease in non-smoke false pixels of an image.

References

[1] Ahmed Fakhir Mutar, dr. Hazim Gati dway “smoke detection based on image processing by using grey and transparency features” Journal of Theoretical and Applied Information Technology 15th November 2018. Vol.96. [2] Ye, S., et al.(2017).\" An effective algorithm to detect both smoke and flame using color and wavelet analysis\". Pattern Recognition and Image Analysis. 27(1): p. 131-138. [3] Kong, S.G., et al.(2016). \"Fast fire flame detection in surveillance video using logistic regression and temporal smoothing\". Fire Safety Journal,. 79: p. 37-43. [4] Shantaiya, S., K. Verma, and K. Mehta.( 2015). \"Multiple object tracking using kalman filter and optical flow\". European Journal of Advances in Engineering and Technology,. 2(2): p. 34-39. [5] Memane, S. and V. Kulkarni. (2015).\" A review on flame and smoke detection techniques in video\'s\". Int. J. of Advanced Research in Electrical, Electronics and Instrumentation Engineering. 4(2): p. 885889 [6] Bosch, A. Serrano, and L. Vergara, “Multisensor Network System for Wildfire Detection Using Infrared Image Processing”, Hindawi Publishing Corporation, the Scientific World Journal, Volume 2013 [7] Avgerinakis, K., A. Briassouli, and I. Kompatsiaris.(2012.). \"Smoke detection using temporal HOGHOF descriptors and energy color statistics from video\". in International Workshop on Multi-Sensor Systems and Networks for Fire Detection and Management. [8] S., P. Piccinini, and R. Cucchiara, Vision based smoke detection system using image energy and color information. Machine Vision and Applications, 2011. 22(4): p. 705-719. [9] Kim, D. and Y.-F. Wang.(2009).\"Smoke detection in video\". in Computer Science and Information Engineering, WRI World Congress on. 2009. IEEE. [10] Yuan, F., A(2008).\"A fast accumulative motion orientation model based on integral image for video smoke detection\".Pattern Recognition Letters. 29(7): p. 925-932

Copyright

Copyright © 2025 Er Harpreet Kaur. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.