UWV-Yolox: A Deep Learning Model for Underwater Video Object Detection

被引:6
|
作者
Pan, Haixia [1 ]
Lan, Jiahua [1 ]
Wang, Hongqiang [1 ]
Li, Yanan [1 ]
Zhang, Meng [1 ]
Ma, Mojie [1 ]
Zhang, Dongdong [1 ]
Zhao, Xiaoran [1 ]
机构
[1] Beihang Univ, Sch Software, Beijing 100191, Peoples R China
关键词
underwater video; object detection; coordinate attention; loss function; frame-level optimization;
D O I
10.3390/s23104859
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
Underwater video object detection is a challenging task due to the poor quality of underwater videos, including blurriness and low contrast. In recent years, Yolo series models have been widely applied to underwater video object detection. However, these models perform poorly for blurry and low-contrast underwater videos. Additionally, they fail to account for the contextual relationships between the frame-level results. To address these challenges, we propose a video object detection model named UWV-Yolox. First, the Contrast Limited Adaptive Histogram Equalization method is used to augment the underwater videos. Then, a new CSP_CA module is proposed by adding Coordinate Attention to the backbone of the model to augment the representations of objects of interest. Next, a new loss function is proposed, including regression and jitter loss. Finally, a frame-level optimization module is proposed to optimize the detection results by utilizing the relationship between neighboring frames in videos, improving the video detection performance. To evaluate the performance of our model, We construct experiments on the UVODD dataset built in the paper, and select mAP@0.5 as the evaluation metric. The mAP@0.5 of the UWV-Yolox model reaches 89.0%, which is 3.2% better than the original Yolox model. Furthermore, compared with other object detection models, the UWV-Yolox model has more stable predictions for objects, and our improvements can be flexibly applied to other models.
引用
收藏
页数:19
相关论文
共 50 条
  • [41] HARNESSING DEEP TRANSFER LEARNING WITH IMAGING TECHNOLOGY FOR UNDERWATER OBJECT DETECTION AND TRACKING IN CONSUMER ELECTRONICS
    Alahmari, Saad
    AL Mazroa, Alanoud
    Mahmood, Khalid
    Alqurni, Jehad saad
    Salama, Ahmed s.
    Alzahrani, Yazeed
    FRACTALS-COMPLEX GEOMETRY PATTERNS AND SCALING IN NATURE AND SOCIETY, 2024,
  • [42] Deep Learning based Object Detection via Style-transferred Underwater Sonar Images
    Lee, Sejin
    Park, Byungjae
    Kim, Ayoung
    IFAC PAPERSONLINE, 2019, 52 (21): : 152 - 155
  • [43] Enhancing Underwater Object Detection Using Advanced Deep Learning De-Noising Techniques
    Umamageswari, Arasakumaran
    Deepa, Sivapatham
    Hussain, Faritha Banu Jahir
    Shanmugam, Padmapriya
    TRAITEMENT DU SIGNAL, 2024, 41 (05) : 2593 - 2602
  • [44] Hybrid deep-learning framework for object-based forgery detection in video
    Tan, Shunquan
    Chen, Baoying
    Zeng, Jishen
    Li, Bin
    Huang, Jiwu
    SIGNAL PROCESSING-IMAGE COMMUNICATION, 2022, 105
  • [45] A Foreign Object Detection Method for Belt Conveyors Based on an Improved YOLOX Model
    Yao, Rongbin
    Qi, Peng
    Hua, Dezheng
    Zhang, Xu
    Lu, He
    Liu, Xinhua
    TECHNOLOGIES, 2023, 11 (05)
  • [46] Advanced deep learning framework for underwater object detection with multibeam forward-looking sonar
    Ge, Liangfu
    Singh, Premjeet
    Sadhu, Ayan
    STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL, 2024,
  • [47] Automatic detection, identification and counting of deep-water snappers on underwater baited video using deep learning
    Baletaud, Florian
    Villon, Sebastien
    Gilbert, Antoine
    Come, Jean-Marie
    Fiat, Sylvie
    Iovan, Corina
    Vigliola, Laurent
    FRONTIERS IN MARINE SCIENCE, 2025, 12
  • [48] Deep active learning for object detection
    Li, Ying
    Fan, Binbin
    Zhang, Weiping
    Ding, Weiping
    Yin, Jianwei
    INFORMATION SCIENCES, 2021, 579 : 418 - 433
  • [49] Learning Deep Relationship for Object Detection
    Xu, Nuo
    Huo, Chunlei
    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2018, E101D (01) : 273 - 276
  • [50] Active Learning for Deep Object Detection
    Brust, Clemens-Alexander
    Kaeding, Christoph
    Denzler, Joachim
    PROCEEDINGS OF THE 14TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS (VISAPP), VOL 5, 2019, : 181 - 190