数字图像模式识别技术在植物叶片识别中的应用  

Application of Digital Image Pattern Recognition Technology to the Identification of Plant Leaves

作　　者：丁常宏[1] 杨月莹[2] 栗明[3] 王振月[1] 

机构地区：[1]黑龙江中医药大学药学院,哈尔滨150040 [2]东北农业大学理学院,哈尔滨150030 [3]黑龙江中医药大学附属第一医院,哈尔滨150040

出　　处：《世界林业研究》2016年第5期17-21,共5页World Forestry Research

基　　金：黑龙江省青年科学技术基金(QC2009C31);黑龙江省教育厅科学技术研究项目(12531019);黑龙江中医药大学科研基金(035128)

摘　　要：植物分类鉴别是森林资源及林下中药资源清查和保护工作的基础,但现在植物的鉴别主要依靠人为进行,效率低、主观性强,且专业人才较少。随着数字信息化技术的发展,构建计算机化的植物模式识别体系已经成为植物鉴别过程中强有力的工具,可为植物学的定种工作提供便捷、准确的保障。文中从叶片特征选择、常用叶片识别方法及图像识别在手机系统上的应用3个方面综述了叶片模式识别的相关研究成果,并探讨了利用计算机视觉辅助植物分类方面的问题和展望,以期为数字图像模式识别技术在植物叶片识别中的应用和进一步研究提供参考。The classification of plants is the basis of inventory and conservation of forest resources and Chinese traditional medicine resources. However, the identification of plants mainly relies on artificial methods, which have however low efficiency, strong subjectivity and few professionals. With high development of digital information technology, the construction of computerized pattern of plant identification system has become a powerful tool in plant identification process, which provides a convenient and accurate guarantee for classification of plant species. This paper summarized the research achievements of leaf pattern recognition from three aspects： the selection of the leaf characteristics, the commonly used methods of leaf identification and the application of image recognition in mobile phone system, and discussed the problems and prospects of computer vision aided plant classification. This study is expected to provide references to the application of digital image pattern recognition technology to the identification of plant leaves as well as the further research in this regard.

关 键 词：植物识别 叶片 模式识别 数字图像 

分 类 号：Q949[生物学—植物学] TP391.41[自动化与计算机技术—计算机应用技术]