Yang Yi, Liu Xiaokun, Li Wenjie, Li Congqiao, Ma Ge, Yang Guangqin, Ren Jing, Ge Siqin
Key Laboratory of Zoological Systematics and Evolution, Institute of Zoology, Chinese Academy of Sciences, Beijing 100101, China.
University of Chinese Academy of Sciences, Beijing 100049, China.
Biology (Basel). 2023 Jul 14;12(7):1006. doi: 10.3390/biology12071006.
Hindwing venation is one of the most important morphological features for the functional and evolutionary analysis of beetles, as it is one of the key features used for the analysis of beetle flight performance and the design of beetle-like flapping wing micro aerial vehicles. However, manual landmark annotation for hindwing morphological analysis is a time-consuming process hindering the development of wing morphology research. In this paper, we present a novel approach for the detection of landmarks on the hindwings of leaf beetles (Coleoptera, Chrysomelidae) using a limited number of samples. The proposed method entails the transfer of a pre-existing model, trained on a large natural image dataset, to the specific domain of leaf beetle hindwings. This is achieved by using a deep high-resolution network as the backbone. The low-stage network parameters are frozen, while the high-stage parameters are re-trained to construct a leaf beetle hindwing landmark detection model. A leaf beetle hindwing landmark dataset was constructed, and the network was trained on varying numbers of randomly selected hindwing samples. The results demonstrate that the average detection normalized mean error for specific landmarks of leaf beetle hindwings (100 samples) remains below 0.02 and only reached 0.045 when using a mere three samples for training. Comparative analyses reveal that the proposed approach out-performs a prevalently used method (i.e., a deep residual network). This study showcases the practicability of employing natural images-specifically, those in ImageNet-for the purpose of pre-training leaf beetle hindwing landmark detection models in particular, providing a promising approach for insect wing venation digitization.
后翅脉序是甲虫功能和进化分析中最重要的形态特征之一，因为它是用于分析甲虫飞行性能和设计类甲虫扑翼微型飞行器的关键特征之一。然而，用于后翅形态分析的手动地标注释是一个耗时的过程，阻碍了翅形态学研究的发展。在本文中，我们提出了一种新颖的方法，使用有限数量的样本检测叶甲（鞘翅目，叶甲科）后翅上的地标。所提出的方法需要将在大型自然图像数据集上训练的预先存在的模型转移到叶甲后翅的特定领域。这是通过使用深度高分辨率网络作为主干来实现的。低阶段网络参数被冻结，而高阶段参数被重新训练以构建叶甲后翅地标检测模型。构建了一个叶甲后翅地标数据集，并在不同数量的随机选择的后翅样本上训练网络。结果表明，叶甲后翅特定地标的平均检测归一化平均误差（100个样本）保持在0.02以下，当仅使用三个样本进行训练时，平均误差仅达到0.045。比较分析表明，所提出的方法优于一种普遍使用的方法（即深度残差网络）。这项研究展示了使用自然图像（特别是ImageNet中的图像）预训练叶甲后翅地标检测模型的实用性，为昆虫翅脉数字化提供了一种有前途的方法。
