摘要
为了实现采摘机器人在复杂的自然场景下正确识别树上果实,来完成果实采摘,研究了不同环境下柑橘的识别方法.针对复杂的自然环境的影响及传统方法的局限性,在可见光和近红外区域择选5个特征波长滤波片,采集得到5幅滤波后的图像,并利用光谱角分类算法完成柑橘识别.试验结果表明,在光照角度、光照强度等不同条件下,柑橘的识别准确度达到96%.研究表明,滤波片光谱图像技术结合光谱角分类算法可以有效地识别自然场景下的成熟柑橘.
In order to guide the robots for harvesting citrus fruit in complex natural scenes, a method based on image processing was developed to identify citrus fruit in the tree canopy. Due to the complexity of natural environment and the deficiency of identification of citrus fruit by the traditional methods, in this work, five narrow band filters in the visible and near-infrared region were selected to capture five images. Spectral angle mapper （SAM） algorithm was attempted to process these images. Experiment results show that citrus can be recognized with an accuracy of 96% in different aspect of illumination angle and intensity. This work demonstrates that spectral imaging technique based on filters combining with SAM algorithm can be used effectively to indentify citrus fruit in complex natural scenes.
出处
《光子学报》
EI
CAS
CSCD
北大核心
2009年第12期3171-3175,共5页
Acta Photonica Sinica
基金
国家高技术研究与发展计划(2006AA10Z263)
国家自然科学基金(30771243)资助
关键词
光谱图像
滤波片
波谱角分类
采摘机器人
识别
柑橘
Spectral imaging
Filter
Spectral angle mapper （SAM）
Harvesting robot
RecognitionCitrus
