Marine animals and deep underwater objects are difficult to recognize and monitor for safety of aquatic life. There is an increasing challenge when the water is saline with granular particles and impurities. In such natural adversarial environment, traditional approaches like CNN start to fail and are expensive to compute. This project involves implementing and evaluating various object detection models, including EfficientDet, YOLOv5, YOLOv8, and Detectron2, on an existing annotated underwater dataset, called the Brackish-Dataset. The dataset comprises annotated image sequences of fish, crabs, starfish, and other aquatic animals captured in Limfjorden water with limited visibility. The aim of this research project is to study the efficiency of newer models on the same dataset and contrast them with the previous results based on accuracy and inference time. Firstly, I compare the results of YOLOv3 (31.10% mean Average Precision (mAP)), YOLOv4 (83.72% mAP), YOLOv5 (97.6%), YOLOv8 (98.20%), EfficientDet (98.56% mAP) and Detectron2 (95.20% mAP) on the same dataset. Secondly, I provide a modified BiSkFPN mechanism (BiFPN neck with skip connections) to perform complex feature fusion in adversarial noise which makes modified EfficientDet robust to perturbations. Third, analyzed the effect on accuracy of EfficientDet (98.63% mAP) and YOLOv5 by adversarial learning (98.04% mAP). Last, I provide class activation map based explanations (CAM) for the two models to promote Explainability in black box models. Overall, the results indicate that modified EfficientDet achieved higher accuracy with five-fold cross validation than the other models with 88.54% IoU of feature maps.

翻译：海洋生物与深层水下目标难以识别和监测，这对水生生物安全构成挑战。当水体含盐且含有颗粒物和杂质时，检测难度进一步增加。在这种天然对抗性环境中，CNN等传统方法开始失效且计算成本高昂。本研究在现有标注水下数据集Brackish-Dataset上实施并评估了多种目标检测模型，包括EfficientDet、YOLOv5、YOLOv8和Detectron2。该数据集包含在能见度有限的利姆峡湾水域中捕获的鱼类、螃蟹、海星及其他水生动物的标注图像序列。本研究旨在探究新一代模型在该数据集上的效率，并与先前基于准确率和推理时间的结果进行对比。首先，我们比较了YOLOv3（平均精度31.10%）、YOLOv4（平均精度83.72%）、YOLOv5（97.6%）、YOLOv8（98.20%）、EfficientDet（平均精度98.56%）和Detectron2（平均精度95.20%）在同一数据集上的结果。其次，我们提出了改进型BiSkFPN机制（带跳跃连接的BiFPN颈部），在对抗性噪声中实现复杂特征融合，使改进型EfficientDet对扰动具有鲁棒性。第三，分析了对抗学习对EfficientDet（平均精度98.63%）和YOLOv5（平均精度98.04%）准确率的影响。最后，为两个模型提供基于类激活映射（CAM）的解释，以促进黑盒模型的可解释性。总体结果表明，改进型EfficientDet通过五折交叉验证实现了比其他模型更高的准确率，其特征图交并比（IoU）达88.54%。