Infrared small target detection based on the combination of single image super-resolution reconstruction and YOLOX

被引：4

作者：

Wang Zhiyong ^{[1
]}

Xiang Xuefu ^{[1
]}

Zeng Kan ^{[1
]}

Zhang Zhenyu ^{[1
]}

Li Yanan ^{[1
]}

Song Dengpan ^{[1
]}

机构：

[1] Automat Res Inst Co Ltd, China South Ind Grp Corp, Mianyang, Sichuan, Peoples R China

来源：

2023 2ND ASIA CONFERENCE ON ALGORITHMS, COMPUTING AND MACHINE LEARNING, CACML 2023 | 2023年

关键词：

Infrared technology; Infrared small target detection; Deep learning; Super-resolution reconstruction; YOLOX;

D O I：

10.1145/3590003.3590104

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

For the infrared search and tracking system, it is necessary to increase the ability to detect small infrared targets against complex backgrounds. YOLOX is a high-performance detector, but its detection performance is constrained when it uses data from low-resolution infrared images with small targets. However, occasionally design constraints and budgetary restraints will prevent the optical system and sensor resolution from being increased enough to improve image quality. Real-ESRGAN is used to solve this issue by reconstructing a high-resolution infrared image from its low-resolution counterpart, which will be used as YOLOX-S's input. Also, the YOLOX-S training strategy is modified further to make it appropriate for the detection of infrared small targets, including the Mosaic and MixUp data augmentation and the size of ground-truth. The average precision achieved by the suggested method in this work increases from 63.70% to 77.19%, which shows a considerable improvement in infrared small target detection when compared with the original model by inputting original images.

引用

页码：547 / 552

页数：6

共 16 条

[1]

Anonymous, 2019, ScienceDB

[2]

Bochkovskiy A, 2020, Arxiv, DOI arXiv:2004.10934

[3]

Ge Z, 2021, Arxiv, DOI [arXiv:2107.08430, DOI 10.48550/ARXIV.2107.08430, 10.48550/arXiv.2107.08430]

[4] Improved Algorithm for Insulator and Its Defect Detection Based on YOLOX [J].

Han, Gujing ;

Li, Tao ;

Li, Qiang ;

Zhao, Feng ;

Zhang, Min ;

Wang, Ruijie ;

Yuan, Qiwei ;

Liu, Kaipei ;

Qin, Liang .

SENSORS, 2022, 22 (16)

[5]

He KM, 2017, IEEE I CONF COMP VIS, P2980, DOI [10.1109/ICCV.2017.322, 10.1109/TPAMI.2018.2844175]

[6]

Jocher G., 2021, YOLOv5

[7] SSD: Single Shot MultiBox Detector [J].

Liu, Wei ;

Anguelov, Dragomir ;

Erhan, Dumitru ;

Szegedy, Christian ;

Reed, Scott ;

Fu, Cheng-Yang ;

Berg, Alexander C. .

COMPUTER VISION - ECCV 2016, PT I, 2016, 9905 :21-37

[8]

Redmon J., 2018, Yolov3: An incremental improvement

[9] Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks [J].

Ren, Shaoqing ;

He, Kaiming ;

Girshick, Ross ;

Sun, Jian .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2017, 39 (06) :1137-1149

[10]

Wang X Y, 2018, Research on infrared dim and small target detection theory and methodology based on sparse dynamic inversion

← 1 2 →