Uncertainty in stock assessment can be reduced if accurate and precise length composition of catch is available. Length data are usually manually collected, although this method is costly and time-consuming. Recently, some studies have estimated fish species and length from images using deep learning by installing camera systems in fishing vessels or a fish auction center (Álvarez -Ellacuria et al., 2020; Lekunberri et al., 2022; Ovalle et al., 2022; Palmer et al., 2022). Once a deep learning model is properly trained, it does not require expensive and time-consuming manual labor. However, several studies on the deep learning models had monitoring fishing practices using electronic monitoring systems; therefore, it is necessary to solve many issues, such as counting the total number of fish in the catch. In this study, we proposed a new deep learning-based method to estimate fish length using images. Species identification was not performed by the model, and images were taken manually by the measurers; however, length composition was obtained only for non-occluded fish detected by the model. A smartphone application was developed to calculate scale information (cm/pixel) from a known size fish box in fish images, and the Mask R-CNN (Region-based convolutional neural networks) model was trained using 76,161 fish to predict non-occluded fish. Two experiments were conducted to confirm whether the proposed method resulted in errors in the length composition. First, we manually measured the total length (TL) for four species and one genus (categories), estimated the TL using a deep learning method, and calculated the bias. Second, multiple fish in a fish box were photographed simultaneously, and the relative difference between the mean TL estimated from the non-occluded fish and the true mean TL from all fish was calculated. The results showed that the biases of all five categories were from −0.69 cm to 0.37 cm and the range of difference was from −1.14 % to 1.40 % regardless of the number of fish in the fish box. The deep learning method was used not to replace the measurer but to increase their measurement efficiency. The proposed method is expected to increase opportunities for the application of deep learning-based fish length estimation in areas of research that are different from the scope of conventional electronic monitoring systems.

中文翻译：

---

使用智能手机应用程序和深度学习方法估计检测为非遮挡的鱼的长度

如果能够获得准确且精确的渔获物长度组成，则可以减少种群评估的不确定性。长度数据通常是手动收集的，尽管这种方法既昂贵又耗时。最近，一些研究通过在渔船或鱼类拍卖中心安装摄像系统，利用深度学习的图像估计鱼类种类和长度（Álvarez -Ellacuria 等人，2020；Lekunberri 等人，2022；Ovalle 等人，2022） ；帕尔默等人，2022）。一旦深度学习模型经过适当的训练，就不需要昂贵且耗时的体力劳动。然而，一些关于深度学习模型的研究使用电子监控系统来监控捕鱼行为；因此，有必要解决许多问题，例如计算捕获的鱼的总数。在这项研究中，我们提出了一种新的基于深度学习的方法来使用图像估计鱼的长度。物种识别不是由模型进行的，图像是由测量人员手动拍摄的；然而，仅获得了模型检测到的非遮挡鱼的长度组成。开发了一款智能手机应用程序，用于计算鱼类图像中已知大小的鱼箱的尺度信息（厘米/像素），并使用 76,161 条鱼训练 Mask R-CNN（基于区域的卷积神经网络）模型来预测非遮挡鱼。进行了两个实验来确认所提出的方法是否会导致长度构成错误。首先，我们手动测量了四个物种和一个属（类别）的总长度（TL），使用深度学习方法估计了TL，并计算了偏差。其次，同时拍摄鱼箱中的多条鱼，并计算非遮挡鱼估计的平均 TL 与所有鱼的真实平均 TL 之间的相对差异。结果显示，无论鱼箱中有多少条鱼，所有五个类别的偏差都在-0.69厘米到0.37厘米之间，差异范围在-1.14%到1.40%之间。深度学习方法的使用并不是为了取代测量者，而是为了提高测量效率。所提出的方法有望增加基于深度学习的鱼长估计在不同于传统电子监测系统范围的研究领域的应用机会。