Research on marine fl exible biological target detection based on improved YOLOv8 algorithm

To address the challenge of suboptimal object detection outcomes stemming from the deformability of marine fl exible biological entities, this study introduces an algorithm tailored for detecting marine fl exible biological targets. Initially, we compiled a dataset comprising marine fl exible biological subjects and developed a Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm, supplemented with a boundary detection enhancement module, to refine fi ne underwater image quality and accentuate the distinction between the images' ' foregrounds and backgrounds. This enhancement mitigates the issue of foreground-background similarity encountered in detecting marine fl exible biological entities. Moreover, the proposed adaptation incorporates a Deformable Convolutional Network (DCN) network module in lieu of the C2f module within the YOLOv8n algorithm framework, thereby augmenting the model's ' s proficiency fi ciency in capturing geometric transformations and concentrating on pivotal areas. The Neck network module is enhanced with the RepBi-PAN architecture, bolstering its capability to amalgamate and emphasize essential characteristics of fl exible biological targets. To advance the model's ' s feature information processing efficiency, fi ciency, we integrated the SimAM attention mechanism. Finally, to diminish the adverse effects of inferior-quality labels within the dataset, we advocate the use of WIoU (Wise-IoU) as a bounding box loss function, which serves to refine fi ne the anchor boxes' ' quality assessment. Simulation experiments show that, in comparison to the conventional YOLOv8n algorithm, our method markedly elevates the precision of marine fl exible biological target detection.