English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

As the treasure house of nature, the ocean contains abundant resources. Butthe coral reefs, which are crucial to the sustainable development of marinelife, are facing a huge crisis because of the existence of COTS and otherorganisms. The protection of society through manual labor is limited andinefficient. The unpredictable nature of the marine environment also makesmanual operations risky. The use of robots for underwater operations has becomea trend. However, the underwater image acquisition has defects such as weaklight, low resolution, and many interferences, while the existing targetdetection algorithms are not effective. Based on this, we propose an underwatertarget detection algorithm based on Attention Improved YOLOv5, calledUTD-Yolov5. It can quickly and efficiently detect COTS, which in turn providesa prerequisite for complex underwater operations. We adjusted the originalnetwork architecture of YOLOv5 in multiple stages, including: replacing theoriginal Backbone with a two-stage cascaded CSP (CSP2); introducing the visualchannel attention mechanism module SE; designing random anchor box similaritycalculation method etc. These operations enable UTD-Yolov5 to detect moreflexibly and capture features more accurately. In order to make the networkmore efficient, we also propose optimization methods such as WBF and iterativerefinement mechanism. This paper conducts a lot of experiments based on theCSIRO dataset [1]. The results show that the average accuracy of our UTD-Yolov5reaches 78.54%, which is a great improvement compared to the baseline.

UTD-Yolov5: A Real-time Underwater Targets Detection Method based on Attention Improved YOLOv5