Title:
Vision Based Technique for Smoke and Fire Detection inMonitored Forest Terrain
Author:
Type:
Co-author(s) :
Uncontrolled Keywords:
fire ; lame ; smoke ; segmentation ; image processing ; classification
Abstract:
Forest fire and/or smoke detection problem is an urgent task as it harms ecology and decreases the overall life quality. In this paper, we examine the problem of fire/smoke detection using images of different parts of the controlled area of the forest. We propose an approach to solve the problem based on full segmentation, simplification and subsequent isolation of segments with and without fire/smoke. A new approach to transform the colored image components which affect the scene brightness to the final results is applied. The procedures of statistical analysis and data classification method are offered. Efficiency of the proposed procedures is shown.
Language:
URL:
Additional Information:
dasat@ipia.sci.am ; samvel1207@gmail.com
Affiliation:
Institute for Informatics and Automation Problems ; Russian-Armenian (Slavonic) University
Country:
Year:
Time period:
Conference title:
10 th International Conference on Computer Science and Information Technologies CSIT 2015