Article citationsMore>>

Qin, D., Leichner, C., Delakis, M., Fornoni, M., Luo, S., Yang, F., et al. (2024) Mobilenetv4: Universal Models for the Mobile Ecosystem. In: Leonardis, A., Ricci, E., Roth, S., Russakovsky, O., Sattler, T. and Varol, G., Eds., Computer Vision—ECCV 2024, Springer, 78-96.
https://doi.org/10.1007/978-3-031-73661-2_5

has been cited by the following article:

• TITLE: Forest Fire Recognition Algorithm Based on Improved RT-DETR

  AUTHORS: Da Mu, Yunfeng Shang, Xinlei Hou

  KEYWORDS: Deep Learning, Forest Fire Prevention, Wildfire Detection, Lightweight

  JOURNAL NAME: Journal of Computer and Communications, Vol.13 No.4, April 30, 2025

  ABSTRACT: In recent years, with the frequent occurrence of global forest fires, fire prevention and control technology has become crucial. The advancement of artificial intelligence technology has provided emerging technical means for forest fire prevention. The RT-DETR model, as a newly emerged large model in recent years, has broken through the NMS limitations of the YOLO series and has shown great potential in the field of image recognition. However, due to its high resource consumption, it is not easy to deploy on embedded devices. To address this issue, a lightweight RT-DETR model has been proposed, which uses MobileNetV4 to optimize the backbone network, reducing parameters, enhancing computational speed, and lowering resource consumption. At the same time, to maintain the performance of forest fire detection, a Hierarchical Resolution Attention mechanism (HLF) and Learnable LPE encoding have been designed to ensure that the model remains lightweight without compromising detection capabilities. This improvement offers a new approach for the application of RT-DETR in the field of forest fire prevention.