Plant Taxonomy-Guided Path-Based Tree Classifier For Large-Scale Plant Species Identification

JOURNAL OF ELECTRONIC IMAGING(2021)

引用 4|浏览11
暂无评分
摘要
Plants play a very important role in human life because they are sources of food, clothes, and medicines. It is a really interesting experience for people to recognize unknown plants when they are wandering on campus or hiking in the wilderness. In the past, only people with professional knowledge could recognize unknown plants by using professional books or comparing them with pre-collected plant specimens.1 Recently, with the increasing power of keyword search engines (i.e., Baidu, Bing, and Google), people can easily get a detailed introduction for specific plant species with only few words that can be used to describe object plant species precisely. But it is still a difficult problem for people without professional training to find appropriate words to describe object plants. With the rapid development of image-based large-scale classification technologies2-15 and deep learning,16-19 it is possible for people to recognize unknown plant species by simply taking a photo of object plant species with only a smart phone. However,A deep learning framework is proposed to recognize large-scale plant species by integrating attention-based deep feature extraction network and plant taxonomy-guided pathbased tree classifier. First, a plant taxonomy is constructed for organizing large-scale fine-grained plant species hierarchically in a coarse-to-fine fashion. Second, a deep learning framework is proposed, where attention mechanism is used to remove useless feature components and a plant taxonomy-guided path-based two-layer tree classifier is used to replace the flat softmax classifier in traditional deep convolutional neural network structure. Furthermore, a specific path-based loss function and back-propagation method are proposed to optimize the weight parameters in both deep network and tree classifier. Experimental results on the Orchid2608 plant dataset can also prove that proposed deep attention network with path-based tree classifier can achieve improvements on large-scale plant species identification task. (c) 2021 SPIE and IS&T [DOI: 10.1117/1.JEI.30 .2.023019]
更多
查看译文
关键词
attention mechanism, plant taxonomy, path-based tree classifier, large-scale plant species identification
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要