一种偏好多目标蜂群算法及其在油茶果图像识别中的应用

被引:4
作者
李昕
李立君
机构
[1] 中南林业科技大学机电工程学院
关键词
机器视觉; 多特征; 蜂群算法;
D O I
暂无
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
针对油茶果采摘机器人机器视觉系统实用性要求,提出了一种基于偏好多目标蜂群算法以解决油茶果目标多特征融合问题。在对油茶果采摘图像进行色差阈值分割后,分别提取分割区域的典型颜色、形态及纹理特征中的八个特征量作为偏好区域对油茶果多特征参数的识别。实验结果表明,使用多特征参数融合方法的识别率较之单特征方法有所提高,在晴天时提高了91.27%,在阴天时提高了94.88%;同时平均识别时间控制在3 500 ms内,达到了油茶果实时采摘的要求,为下一步在智能油茶采摘机器人中的应用打下了基础。
引用
收藏
页码:4779 / 4781+4785 +4785
页数:4
相关论文
共 20 条
[1]   不同生长状态下多目标番茄图像的自动分割方法 [J].
尹建军 ;
毛罕平 ;
王新忠 ;
陈树人 ;
张际先 .
农业工程学报, 2006, (10) :149-153
[2]  
Robotic harvesting system for eggplants. Shigehiko Hayashi,Karsunobu Ganno,Yukitsugu Ishii,et al. JATQ . 2002
[3]  
Detection of citrus canker u-sing hyperspectral reflectance imaging with spectral informa-tion divergence. Jianwei Qin,Thomas F Burks. Journal of Food engineering . 2009
[4]  
Detection of red ripe toma-toes on stem using image processing techniques. HOSNA M M,REZA A,MAHMOUD O. Journal of American Science . 2011
[5]  
Recognition and classifi-cation of external skin damage in citrus fruit using multispectral data and morphological features. BLASCO J,ALEIXOS N,GOMEZ S,et al. Biosystems Engineering . 2009
[6]  
A threshold selection method from gray-level histograms. Nobuyuki Otsu. IEEE Transactions on Systems Man and Cybernetics . 1979
[7]   Recognition and classification of external skin damage in citrus fruits using multispectral data and morphological features [J].
Blasco, J. ;
Aleixos, N. ;
Gomez-Sanchis, J. ;
Molto, E. .
BIOSYSTEMS ENGINEERING, 2009, 103 (02) :137-145
[8]  
An idea based on honey bee swarm for nu-merical optimization. Karaboga D. Technical Report-TR06 Oc-tober Erciyes University . 2005
[9]  
On-tree fruit recognition using tex-ture properties and color data. Zhao J,Tow J,Katupitiya J. 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems . 2005
[10]   Detecting fruits in natural scenes by using spatial-frequency based texture analysis and multiview geometry [J].
Rakun, J. ;
Stajnko, D. ;
Zazula, D. .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2011, 76 (01) :80-88