da Silva Arthur Gonsales, de Oliveira Roger Pinho, de Oliveira Bastos Caio, de Carvalho Elena Almeida, Gomes Bruno Duarte
Departamento de Ciência de Dados, Instituto Tecnológico Vale, Belém, Brazil.
Centro de Ciências Biológicas e da Saúde, Universidade da Amazônia, Belém, Brazil.
Front Artif Intell. 2025 Aug 12;8:1524380. doi: 10.3389/frai.2025.1524380. eCollection 2025.
Image classification is a highly significant field in machine learning (ML), especially when applied to address longstanding and challenging issues in the biological sciences, such as specie recognition and biodiversity conservation. In this study, we present the development of a hybrid machine learning-based tool suitable for deployment on mobile devices. This tool is aimed at processing and classifying three-dimensional samples of endemic lizard species from the Amazon rainforest. The dataset used in our experiment was collected at the Museu Paraense Emílio Goeldi (MPEG), Belém-PA, Brazil, and comprises three species: (a) ; (b) ; and (c) . We compared the effectiveness of four artificial neural networks (ANN) for feature extraction: (a) MobileNet; (b) MobileNetV2; (c) MobileNetV3-Small; and (d) MobileNetV3-Large. Additionally, we evaluated five classical ML models for classifying the extracted patterns: (a) Support Vector Machine (SVM); (b) GaussianNB (GNB); (c) AdaBoost (ADB); (d) K-Nearest Neighbors (KNN); and (e) Random Forest (RF). The performance metrics of all classifiers were very close, we used the McNemar's test on each model's confusion matrix to evaluate and compare their statistical significance. Our best model was a combination of a 2.9 million parameters MobileNetV3-Small as the feature extractor, with a linear kernel-based SVM as the classifier, which achieved accuracy of 0.955, precision of 0.948, recall of 0.948, and f1-score of 0.948. The results indicated that the use of a small deep learning (DL) model, in combination with a classical ML algorithm, emerges as a viable technique for classifying three-dimensional representations of lizard species samples. Such an approach facilitates taxonomic identification work for professionals in the field and provides a tool adaptable for integration into mobile data recording equipment, such as smartphones, and benefiting from more morphological features extracted from three-dimensional samples instead of two-dimensional images.
图像分类是机器学习（ML）中一个非常重要的领域，特别是在应用于解决生物科学中长期存在的具有挑战性的问题时，例如物种识别和生物多样性保护。在本研究中，我们展示了一种基于混合机器学习的工具的开发，该工具适合在移动设备上部署。此工具旨在处理和分类来自亚马逊雨林的特有蜥蜴物种的三维样本。我们实验中使用的数据集是在巴西帕拉州贝伦的帕拉恩塞博物馆埃米利奥·戈尔迪博物馆（MPEG）收集的，包括三个物种：（a） ；（b） ；和（c） 。我们比较了四种用于特征提取的人工神经网络（ANN）的有效性：（a）MobileNet；（b）MobileNetV2；（c）MobileNetV3 - 小；和（d）MobileNetV3 - 大。此外，我们评估了五种用于对提取模式进行分类的经典机器学习模型：（a）支持向量机（SVM）；（b）高斯朴素贝叶斯（GNB）；（c）自适应增强（ADB）；（d）K近邻（KNN）；和（e）随机森林（RF）。所有分类器的性能指标非常接近，我们对每个模型的混淆矩阵使用麦克尼马尔检验来评估和比较它们的统计显著性。我们最好的模型是一个具有290万个参数的MobileNetV3 - 小作为特征提取器，与一个基于线性核的支持向量机作为分类器的组合，其准确率为0.955，精确率为0.948，召回率为0.948，F1分数为0.948。结果表明，使用小型深度学习（DL）模型与经典机器学习算法相结合，是对蜥蜴物种样本的三维表示进行分类的一种可行技术。这种方法便于该领域专业人员进行分类鉴定工作，并提供了一种适用于集成到移动数据记录设备（如智能手机）中的工具，且受益于从三维样本而非二维图像中提取的更多形态特征。
