YOLO-FIRI: Improved YOLOv5 for Infrared Image Object Detection
Autor: | Li Shasha, Xiaorong Xu, Yongjun Li, Yao Li, Mengjun Li |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2021 |
Předmět: |
Image fusion
General Computer Science Computer science business.industry Feature extraction Detector General Engineering Attention mechanism Pattern recognition object detection Convolutional neural network image fusion Object detection Constant false alarm rate TK1-9971 YOLOv5 General Materials Science Noise (video) Artificial intelligence Electrical engineering. Electronics. Nuclear engineering infrared image Focus (optics) business |
Zdroj: | IEEE Access, Vol 9, Pp 141861-141875 (2021) |
ISSN: | 2169-3536 |
Popis: | To solve object detection issues in infrared images, such as a low recognition rate and a high false alarm rate caused by long distances, weak energy, and low resolution, we propose a region-free object detector named YOLO-FIR for infrared (IR) images with YOLOv5 core by compressing channels, optimizing parameters, etc. An improved infrared image object detection network, YOLO-FIRI, is further developed. Specifically, while designing the feature extraction network, the cross-stage-partial-connections (CSP) module in the shallow layer is expanded and iterated to maximize the use of shallow features. In addition, an improved attention module is introduced in residual blocks to focus on objects and suppress background. Moreover, multiscale detection is added to improve small object detection accuracy. Experimental results on the KAIST and FLIR datasets show that YOLO-FIRI demonstrates a qualitative improvement compared with the state-of-the-art detectors. Compared with YOLOv4, the mean average precision (mAP50) of YOLO-FIRI is increased by 21% on the KAIST dataset, the speed is reduced by 62%, the parameters are decreased by 89%, the weight size is reduced by more than 94%, and the computational costs are reduced by 84%. Compared with YOLO-FIR, YOLO-FIRI has an approximately 5% to 20% improvement in AP, AR (average recall), mAP50, F1, and mAP50:75. Furthermore, due to the shortcomings of high noise and weak features, image fusion can be applied to image preprocessing as a data enhancement method by fusing visible and infrared images based on a convolutional neural network. |
Databáze: | OpenAIRE |
Externí odkaz: |