Few-Shot Fine-Grained Forest Fire Smoke Recognition Based on Metric Learning

Abstract

To date, most existing forest fire smoke detection methods rely on coarse-grained identification, which only distinguishes between smoke and non-smoke. Thus, non-fire smoke and fire smoke are treated the same in these methods, resulting in false alarms within the smoke classes. The fine-grained identification of smoke which can identify differences between non-fire and fire smoke is of great significance for accurate forest fire monitoring; however, it requires a large database. In this paper, for the first time, we combine fine-grained smoke recognition with the few-shot technique using metric learning to identify fire smoke with the limited available database. The experimental comparison and analysis show that the new method developed has good performance in the structure of the feature extraction network and the training method, with an accuracy of 93.75% for fire smoke identification.

Design of smoke detection system using deep learning and sensor fusion with recursive feature elimination cross-validation

“…Smoke sensors that are commonly used have limitations on the level of sensitivity so that there is a minimum volume of smoke to be detected as a fire; ionization-based smoke sensors have limits on the amount of smoke detected so that they are less effective in early detection of smoldering fires, and on photoelectric-based smoke, sensors have limitations on the detection activation time and the type of smoke that passes through it [3]. Futhermore, smoke sensors often give false positives when detecting a fire, such as when cooking steam is cooking and dust when cleaning [4]. Due to this limitation, it is necessary to have a more accurate and precise sensosr for detecting the presence of smoke as a fire prevention measure.…”