Hanson Nora N, Ounsley James P, Henry Jason, Terzić Kasim, Caneco Bruno
Freshwater Fisheries Laboratory, Marine Directorate, Scottish Government, Pitlochry PH16 5LB, United Kingdom.
School of Computer Science, University of St Andrews, St Andrews KY16 9SX, United Kingdom.
Biol Methods Protoc. 2024 Jul 31;9(1):bpae056. doi: 10.1093/biomethods/bpae056. eCollection 2024.
Teleost fish scales form distinct growth rings deposited in proportion to somatic growth in length, and are routinely used in fish ageing and growth analyses. Extraction of incremental growth data from scales is labour intensive. We present a fully automated method to retrieve this data from fish scale images using Convolutional Neural Networks (CNNs). Our pipeline of two CNNs automatically detects the centre of the scale and individual growth rings (circuli) along multiple radial transect emanating from the centre. The focus detector was trained on 725 scale images and achieved an average precision of 99%; the circuli detector was trained on 40 678 circuli annotations and achieved an average precision of 95.1%. Circuli detections were made with less confidence in the freshwater zone of the scale image where the growth bands are most narrowly spaced. However, the performance of the circuli detector was similar to that of another human labeller, highlighting the inherent ambiguity of the labelling process. The system predicts the location of scale growth rings rapidly and with high accuracy, enabling the calculation of spacings and thereby growth inferences from salmon scales. The success of our method suggests its potential for expansion to other species.
硬骨鱼的鳞片会形成与体长的体细胞生长成比例的独特生长环，并且通常用于鱼类年龄和生长分析。从鳞片中提取增量生长数据需要耗费大量人力。我们提出了一种使用卷积神经网络（CNN）从鱼鳞图像中检索此数据的全自动方法。我们由两个CNN组成的管道会自动检测鳞片的中心以及沿着从中心发出的多个径向断面的各个生长环（环片）。焦点检测器在725张鳞片图像上进行了训练，平均精度达到99%；环片检测器在40678个环片标注上进行了训练，平均精度达到95.1%。在鳞片图像中生长带间距最窄的淡水区域，对环片的检测置信度较低。然而，环片检测器的性能与另一位人工标注者的性能相似，这突出了标注过程中固有的模糊性。该系统能够快速且高精度地预测鳞片生长环的位置，从而能够计算间距，进而从鲑鱼鳞片推断生长情况。我们方法的成功表明了其扩展到其他物种的潜力。
