Medicinal plants are gaining attention in the pharmaceutical industry due to having less harmful effects reactions and cheaper than modern medicine. Based on these facts, many researchers have shown considerable interest in the research of automatic medicinal plants recognition. There are various opportunities for advancement in producing a robust classifier that has the ability to classify medicinal plants accurately in real-time. In this paper, various effective and reliable machine learning algorithms for plant classifications using leaf images that have been used in recent years are reviewed. The review includes the image processing methods used to detect leaf and extract important leaf features for some machine learning classifiers. These machine learning classifiers are categorised according to their performance when classifying leaf images based on typical plant features, namely shape, vein, texture and a combination of multiple features. The leaf databases that are publicly available for automatic plants recognition are reviewed as well and we conclude with a discussion of prominent ongoing research and opportunities for enhancement in this area.

中文翻译：

---

药用植物识别中的机器学习：综述

药用植物因其有害反应较少且比现代药物便宜而在制药业中受到关注。基于这些事实，许多研究人员对药用植物自动识别的研究表现出相当大的兴趣。在生产能够实时准确地对药用植物进行分类的强大分类器方面，存在各种进步机会。在本文中，回顾了近年来使用的各种有效且可靠的使用叶子图像进行植物分类的机器学习算法。该评论包括用于检测叶子和为某些机器学习分类器提取重要叶子特征的图像处理方法。这些机器学习分类器根据它们在基于典型植物特征（即形状、静脉、纹理和多个特征的组合）对叶子图像进行分类时的性能进行分类。公开可用于自动植物识别的叶数据库也进行了审查，最后我们讨论了该领域正在进行的突出研究和改进机会。