Convolutional Neural Networks based Fire Detection in Surveillance Videos

Convolutional Neural Networks based Fire Detection in Surveillance Videos

Mr. S. Nelson, M.E., (Author) Assistant Professor,

Department of Information Technology ,

V.S.B. Engineering College, Karur. Tamilnadu, India.

Ms. S. Suganya, B.Tech,

Student, Department of Information Technology,

V.S.B. Engineering College, Karur, Tamilnadu, India.

Ms. M. Selvabharathi, B.Tech., Student, Department of Information Technology,

V.S.B. Engineering College, Karur. Tamilnadu, India.

Ms. C. Vishnupriya, B.Tech., Student, Department of Information Technology ,

V.S.B. Engineering College, Karur, Tamilnadu, India.

Abstract:- In this project is to study the detection of flames in a video byusing motion and edge detection technique. This is an improved method over all the existing ones. This method detects the edges of the flames properly by removing the noises in the flames. This paper focus is on identifying gray cycle pixels for the detection of flame. Its optimizing technique to detection of the flame, which generated because of smoke and of spreading of fire pixel and the area spread of flame. These systems can be used to reduce false detection fire. The novel system simulate the existing fire detection techniques with above given new techniques of fire detection and give optimized way to detect the fire in terms of less false detection of fire by give the accurate result of fire occurrence. The strength fire detection is the ability to monitor indoor and outdoor applications by using videos. The novel system also gives the opportunity to adjust the system by applying different combination of fire detecting techniques which will help in implementation of system according to different sensitive area requirement.

KeywordsComponent; AI techniques; storing massive knowledge; Adaboos; tregression

I.INTRODUCTION.

Fire detection systems are a critical element of any building design. For high-rise buildings and multi-winged structures such as hospitals and hotels, these designs can become complex

.Automatic fire detection systems, when combined with other elements of an emergency response and evacuation plan, can significantly reduce property damage, personal injuries, and loss of life from fire in the workplace. Their main function is to quickly identify a developing fire and alert building occupants and emergency response personnel before extensive damage occurs. Automatic fire detection systems do this by using electronic sensors dynamical patterns and construct drifts. in an exceedingly traffic context, construct drifts square measure the changes to the distributions {of knowledge| of knowledge |of information} in an exceedingly traffic data stream over time

. supported the character of fluctuations in knowledge streams, these changes square measure more classified as continual and non-recurrent construct drifts. as an example, traffic jam changes thanks to peak/off-peak traffic square measure a continual construct drift whereas associate

accident or breakdown could be a non-recurrent construct drift. Special importance ought to be placed into distinctive non continual construct drifts because it may have an effect on the whole road network. Fire detection systems are a critical element of building design.

For high-rise buildings and multi-winged structures such as hospitals and hotels, these designs can become complex. Automatic fire detection systems, when combined with other elements of an emergency response and evacuation plan, can significantly reduce property damage, personal injuries, and loss of life from fire in the workplace. Their main function is to quickly identify a developing fire and alert building occupants and emergency response personnel before extensive damage occurs. Automatic fire detection systems do this by using electronic sensors

1. EXISTING SYSTEM

   The pixel-level methods are fast due to usage of pixel-wise features such as colors and flickers; however, their performance is not attractive as such methods can be easily biased. Compared to pixel-level methods, blob-level flame detection methods show better performance as such methods consider blob-level candidates for features extraction to detect flame. The major problem with such methods is the difficulty in training their classifiers due to numerous shapes of fire blobs. To improve the accuracy, researchers attempted to explore color and motion features for flame detection.

   Each pixel has a particular color; that color is described by the amount of red, green and blue in it. If each of these components has a range 0255, this gives a total of 2563different possible colors. Such an image is a stack of three matrices; representing the red, green and blue values for each pixel. This means that for every pixel there correspond 3 values.

2. DISADVANTAGES OF EXISTING SYSTEM

   • The development of time period machine learning algorithms and prediction schemes for non- recurrent fire incidents that impact a whole road network

   • A majority of existing approaches target

     freeways and highways, with terribly restricted attention to blood vessel networks because of the technical challenges of integration multiple streams of fire information, so fails once crucial fire propagation within the entire network.

   • An image processor does the functions of image acquisition, storage, preprocessing, segmentation, representation, recognition and interpretation and finally displays or records the resulting image

3. PROPOSED SYSTEM

   In this work, we have a tendency to distinguish on-line learning and progressive learning. on-line learning updates the model exploitation every incoming information that arrives throughout the operation, while not storing. As such, on-line learning is used to handle massive volumes of streaming knowledge inward at high rate. progressive learning is learning from batches information |of information} at distinct time intervals and has the potential to stabilize the historical knowledge of the educational model over novel learning. Hence, the model is updated to any new information that's received whereas keeping its existing data intact. Further, it's essential that non- recurrent conception drifts square measure known and utilized for updated traffic propagation and traffic flow prediction models during a time period manner.

   To this finish, we have a tendency to any address many key considerations that square measure underexplored in current ITS, to support the event of a holistic fire detection management. Following an in depth review of current literature in ITS we have a tendency to known the subsequent four current challenges that haven't been sufficiently self-addressed.

   1. the event of time period machine learning algorithms and prediction schemes for fire detection using surveillance video.

   2. The operator console consists of equipment and arrangements for verification of intermediate results and for alterations in the software as and when require. The operator is also capable of checking for any resulting errors and for the entry of requisite data.

   3. Image enhancement operations improve the qualities of an image like improving the images contrast and brightness characteristics, reducing its noise content, or sharpen the details. This just enhances the image and reveals the same information in more understandable image. It does not add any information to it. Each pixel has a particular color; that color is described by the amount of red, green and blue in it. If each of thse components has a range 0255, this gives a total of 2563different possible colors. Such an image is a stack of three matrices; representing the red, green and blue values for each pixel. This means that for every pixel there correspond 3 values. Our proposed system provides fire detection using a

      simple algorithm. It is as described below. Firstly, the image frame is acquired from the live video feed.

   4. The RGB color model is then applied to the frame. The resultant RGB frame is then converted to a HSV frame. This frame is then passed through thresholding, median blurring (to remove noise), Background Subtraction, Sobel edge detection, and motion detection windows.

   5. A suitable response is displayed on the window monitor, and an alarm buzzer is sounded.

4. ADVANTAGES OF PROPOSED SYSTEM

5. FEASIBILITY STUDY

   The feasibility study deals with all the analysis that takes up in developing the project. Each structure has to be thought of in the developing of the project, as it has to serve the end user in a user- friendly manner. One must know the type of information to be gathered and the system analysis consist of collecting, Organizing and evaluating facts about a system and its environment.

   The main objective of the system analysis is to study the existing operation and to learn and accomplish the processing activities. Calculating cloud area status at a given refresh period through windows application needs to be analyzed well. Cloud areas must be grouped based on their processing ability. According to their processing and storage power, the partial job needs to assign to them. The details are processed through coding themselves. It will be controlled by the programs alone.

   1. ECONOMIC FEASIBILITY

      The organization has to buy a personal computer with a keyboard and a mouse, this is a direct cost. There are many direct benefits of covering the manual system to computerized system. The user can be given responses on asking questions, justification of any capital outlay is that it will reduce expenditure or improve the quality of service or goods, which in turn may be expected to provide the increased profits.

   2. OPERATIONAL FEASIBILITY

      The Proposed system accessing process to solves problems what occurred in existing system. The current day-to-day operations of the organization can be fit into this system. Mainly operational feasibility should include on analysis of how the proposed system will affects the organizational structures and procedures.

   3. TECHNICAL FEASIBILITY

   The cost and benefit analysis may be concluded that computerized system is benefit in todays fast moving world. The assessment of technical feasibility must be based on an outline design of the system requirements in terms of input, output, files, programs and procedure. The project aims to assign multiple nodes after the job is split according to the nodes capability from the given

   application. The current system aims to overcome the problems of the existing system. The current system is to reduce the technical skill requirements so that more number of users can access the application.

6. PROJECT DESCRIPTION

   Module

   1. Open cv.

   2. Matplotlib.

   3. Numpy.

   4. GaussianBlur.

   1. Open CV

      Open CV supports some models from deep learning frameworks like Tensor Flow, Torch, Py Torch (after

      converting to an ONNX model)and Caffe according to a defined list of supported layers. It promotes Open Vision Capsules which is a portable format, compatible with all other formats.

      • Cost effective

      • High Efficiency

      • When the controller receives the fire detected Information, buzzer goes on for indication purpose.

      • This system installed with the GSM. It is used to send the alert message.

   2. Matplotlib

      Matplotlib is a plotting library for the Python programming language and its numerical mathematics extension NumPy. It provides an object-oriented API for embedding plots into applications using general-purpose GUI

      toolkits like Tkinter, wxPython, Qt, or GTK+. There is also a procedural "pylab" interface based on a state

      machine (like OpenGL), designed to closely resemble that of MATLAB, though its use is discouraged. SciPy makes use of Matplotlib.

   3. NumPy

      NumPy (pronounced /nmpa/ (NUM-py) or sometimes /nmpi/(NUM-pee)) is a library for the Python programming language, adding support for large, multi- dimensional arrays and matrices,

      along with a large collection

      of high-level mathematical functions to operate on these arrays. The ancestor of NumPy, Numeric, was originally created by Jim Hugunin with contributions from several other developers. In

      2005, Travis Oliphant created NumPy by incorporating features of the competing Numarray into Numeric, with extensive modifications. NumPy is open-source software and has many contributors.

   4. Gaussian blur

   In image processing, a Gaussian blur (also known as Gaussian smoothing) is the result of blurring an image by a Gaussian function (named after mathematician and scientist Carl Friedrich Gauss). It is a widely used effect in graphics software, typically to reduce image noise and reduce detail. The visual effect of this blurring technique is a smooth blur resembling that of viewing the image through a translucent screen, distinctly different from

   the bokeh effect produced by an out-of-focus lens or the shadow of an object under usual illumination. Gaussian smoothing is also used as a pre-processing stage in computer vision algorithms in order to enhance image structures at different scalessee scale space representation and scale space implementation.

7. SYSTEM ARCHITECTURE

8. Conclusion

   In our project we propose a fire detection algorithm which is free from sensors as the ordinary fire detection systems contain. The objective of this project was to create a system which would be able to detect fire as early as possible from a live video feed. System is expected to detect fire while it is still small and has not grown to mammoth proportions. Also, the hardware is minimal and has been already existent in places, thus saving capital. It also saves cost by getting rid of expensive temperature and heat sensors etc. Based on the results produced, the system has proven to be effective at detecting fire. This system is an amalgamation of various fire detection algorithms.

   The system can be made weather proof Smoke detection along with fire detection can be added as a feature System Optimization and Delay Reduction i.e. Lesser latency may be achieved. System can be used to detect forest fires and may be embedded on a drone or any other UAV for surveillance purposes of property. The system can have military applications. The system can be used for rescue operations on land and in sea.

9. ACKNOWLEDGMENT

   Authors would like to thank VicRoadsVictorian Road Authority, Australia for their consultations in domain knowledge and providing us with real road traffic data. Authors would also like to acknowledge the in-kind support from the Data to Decisions Cooperative Research Centre (D2D CRC) as part of their analytics and decision support Program

10. REFERENCE

1. Sonali K. M., Poojashree M., Shilpashree A., Sindhu N., Rekha

   K. S.,Automated Fire Detection Surveillance System, CSE, NIE, Mysuru,International Journal of Current Trends in Engineering & Research

2. OpenCV Documentation (Background Subtraction) tutorial_py_bg_subtracti on.html

3. P. KadewTraKuPong, R. Bowden, An improved adaptive background mixture model for real-time tracking with shadow detection 2001

4. Z. Zivkovic, Improved adaptive Gaussian mixture model for background subtraction 2004

5. Z. Zivkovic,Efficient Adaptive Density Estimation per Image Pixel for the Task of Background Subtraction 2006 [6] OpenCV Documentation (Contour analysis) http://docs.opencv.org/trunk/d4/d73/tutorial_py_contours_be gin.html

1. OpenCV Documentation (Structural analysis) http://docs.opencv.org/2.4/modules/imgproc/doc/structural_a nalysis_and_shape_descriptors.html

2. S. Bharathi, S. Gokilapriya, N. Elango& P. Vidhya, fire detection and fire signature using color models for security, International Journal ofCurrent Research and Modern Education (IJCRME) Special Issue, NCFTCCPS 2016

3. Akshay Thokale , Poonam Sonar, Review on Vision Based Fire Flame Detection, International Journal of Innovative Research in Science, Engineering and Technology, Vol. 4, Issue 9, September 2015

4. TurgayCelik, Hasan Demirel, HuseyinOzkaramanli, and Mustafa Uyguroglu, "Fire detection using statistical color model in video sequences," J. Vis. Comun. Image Represent., vol. 18, pp. 176185, April 2007

5. TurgayCelik, "Fast and Efficient Method for Fire Detection Using Image Processing," ETRI Journal, Volume 32, Number 6,

   December 2010

6. Rafael C. Gonzalez and Richard E. Woods: Digital Image Processing, 2009

7. Raspberry Pi official website http://www.raspberrypi.org

8. OpenCV official website http://www.opencv.org International Journal of Engineering Research & Technology (IJERT) ISSN: 2278-0181http:// www.ijert.org