Fire detection for early fire alarm based on optical flow video processing | |
| Author | Suchet Rinsurongkawong |
| Call Number | AIT Thesis no.ISE-12-34 |
| Subject(s) | Fire alarms Fire detectors Image processing |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Mechatronics, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. ISE-12-34 |
| Abstract | Conventional fire detection systems are generally limited to indoors and they ha ve slow response for detect fire. A system that is able to detect fire by processing real - time video images by using camera would both work in open area and it doesn't need a high budget. It can also be used in accordance with high resolution cameras for better performance. This th esis mainly focuses on design the algorithm for early fire detection based on the Lucas - Kanade optical flow algorithm that is able to detect fire in real time in a video stream from a monocular camera. The method works both indoors and in open areas. It detects fire at the beginning of the burning process, enabling an earlier response than would be possible with a conventional fire detector. The method performs background subtraction to identify moving pixels in the scene then filters for colors consistent with fire. Growth rate analysis is applied to the contours identified in the previous step. Fi nally, the use of a Lucas - Kanade optical flow pyramid in the regions identified in the growth analysis step and analysis of the variance of the optical flow feature points’ motion is purposed to find the flow rate . In an experiment , comparing the growth ra te method and the flow rate method alone with the combination of growth rate and optical flow analysis, the results show that the combination algorithm improves the accuracy of fire detection for early fire alarm and reduce false alarm s by solving the fire - like color and non - fire objects. |
| Year | 2012 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis ; no. ISE-12-34 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Industrial Systems Engineering (DISE) |
| Academic Program/FoS | Industrial Systems Engineering (ISE) |
| Chairperson(s) | Manukid Parnichkun |
| Examination Committee(s) | Dailey, Mathew N.;Tarin Tantimuratha;Mongkol Ekpanyapong |
| Scholarship Donor(s) | Thailand (HM King) |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2012 |