文献类型：期刊文献

英文题名：Pixel-level automatic annotation for forest fire image

作者：Yang, Xubing[1] Chen, Run[1] Zhang, Fuquan[1] Zhang, Li[1,2] Fan, Xijian[1] Ye, Qiaolin[1] Fu, Liyong[3]

第一作者：Yang, Xubing

通信作者：Yang, XB[1];Zhang, L[1];Zhang, L[2]

机构：[1]Nanjing Forestry Univ, Coll Informat Sci & Technol, Nanjing 210037, Peoples R China;[2]Nanjing Univ Aeronaut & Astronaut, MIIT Key Lab Pattern Anal & Machine Intelligence, Nanjing 210016, Peoples R China;[3]Chinese Acad Forestry, Inst Forest Resource Informat Tech, Beijing 100091, Peoples R China

年份：2021

卷号：104

外文期刊名：ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE

收录：;EI(收录号：20212610567894);Scopus(收录号：2-s2.0-85108634012);WOS:【SCI-EXPANDED(收录号:WOS:000694881200027)】；

基金：We would thank Dr. Pramod Vemulapalli and his group at Simon Fraiser University, for their kd-tree code. This research was supported in part by the Central Public-interest Scientific Institution Basal Research Fund, China (Grant No. CAFYBB2019QD003), Natural Science Foundation of China under Grant 31670554 and 61802193, the Fundamental Research Funds for the Central Universities, China (NJ2020023), and the Jiangsu Science Foundation, China under Grant BK20170934.

语种：英文

外文关键词：Fire detection; Convex hull; Pixel-level; Image annotation

摘要：We propose an automatic annotation method for forest fire images in the level of pixel, where supervise information is introduced by interactive convex hulls. Instead of usual rectangle-/regular-shaped regions, we propose a convex hull algorithm for visually selecting polygonal (irregular) fire and no-fire regions. Guided by the goals of forest fire monitoring systems: high fire detection rate (true-positive) and then low false alarm rate (false-positive), we construct a k-nearest neighbor (kNN) based KD-tree to speed annotation. Compared to state-of-the-art, the proposed method not only widens the view of fire detection from conventional two-class to multi-class classification problem to meet complex forest image background, but also relaxes the limit of i.i.d (independent and identical distribution) hypothesis on machine learning methods. Furthermore, it is simple to use, which just relies on pixel information and avoids considering additional auxiliary features from multiple color spaces. Experimental evaluations are carrying on forest fire images, MIVIA dead-directional videos, and more challenging omni-directional videos. The comparison demonstrates that the proposed pixel-level annotation method is able to achieve higher fire detection rate and lower false alarm rate at the same time.