ZHANG Wanhong,ZHU Yuanjun.Study on Image Segmentation for Bagged Apple[J].Northern Horticulture,2017,41(08):6-9.[doi:10.11937/bfyy.201708002]
套袋苹果影像的分割方法研究
- Title:
- Study on Image Segmentation for Bagged Apple
- 关键词:
- 套袋苹果; R、G和B值分布特征; 影像分割
- 文献标志码:
- A
- 摘要:
- 为了准确识别并分割套袋苹果，对影像中的套袋苹果及其背景在RGB颜色空间模式下，分析各目标物的R（红）、G（绿）及B（蓝）值分布特征，根据各目标物所呈现的R、G及B值差异选取适合于套袋苹果的分割条件，对套袋苹果进行分割，将分割结果与多阈值分割法以及K均值聚类法分割结果进行比较。结果表明：分割法优于以上2种方法，同时，根据该试验提出的分割方法计算苹果识别率，识别率可达93.4%，表明该试验算法对套袋苹果的分割准确率高。
- Abstract:
- In order to accurately segment of bagged apple，the feature of R，G and B values of RGB mode for bagged apple and other targets image was analyzed，and then the bagged apples were segmented based on the difference of R，G and B values of each target.Compared the results of image segmentation obtained by R，G and B distribution with the results obtained by the multiple threshold and K-means clustering analysis.The results indicated that the image segmentation method based on the R，G and B distribution was better than multiple threshold and K-means clustering analysis.Meanwhile，computing the identification rate of apple on the basis of image segmentation using the method of R，G and B distribution，the results showed that the identification rate was 93.4%.Consequently，it could be concluded that the image segmentation with the help of R，G and B distribution was reliable.
参考文献/References:
[1]王海波,李林光,陈学森,等.中早熟苹果品种果实的风味物质和风味品质[J].中国农业科学,2010(11):2300-2306. [2]刘天军,蔡起华,朱玉春.气候变化对苹果主产区产量的影响：来自陕西省6个苹果生产基地县210户果农的数据[J].中国农村经济,2012(5):32-40. [3]聂继云,吕德国,李静,等.苹果果实中类黄酮化合物的研究进展[J].园艺学报,2009(9):1390-1397. [4]唐周怀,陈川,惠伟,等.套袋苹果黑点病的发生规律[J].西北农林科技大学学报(自然科学版),2003(2):59-61. [5]吕德国,陈军,高文胜,等.套袋苹果不同纸袋内不同时期真菌种群结构研究[J].沈阳农业大学学报,2009(1):80-83. [6]张春龙,张楫,张俊雄,等.近色背景中树上绿色苹果识别方法[J].农业机械学报,2014(10):277-281. [7]牛希泉,梁艳梅.自然场景下成熟苹果彩色图像分割方法的研究[J].光电子激光,2007(12):1453-1456. [8]胡婵莉,赵德安,赵宇艳,等.基于极值的重叠苹果识别方法研究[J].农机化研究,2016(3):42-46. [9]HEINEMANN P H，VARGHESE Z A，MORROW C T，et al.Machine vision inspection of ‘Golden Delicious’ apples[J].Applied Engineering in Agriculture,1995,11(6):901-906. [10]司永胜,乔军,刘刚,等.基于机器视觉的苹果识别和形状特征提取[J].农业机械学报，2009(8):161-165,73. [11]李宏利,何东健.图像中被遮挡苹果的还原及定位技术研究[J].农机化研究,2013(9):20-23. [12]SELIM S Z，ISMAIL M A.K-means type algorithms:a generalized convergence theorem and characterization of local optimality[J].IEEE Trans Pattern Analysis Machine Intelligence,1984,PAMI-6(1):81-87.
备注/Memo
第一作者简介：张万红（1974-），男，陕西扶风人，博士，助理研究员，研究方向为生态模型及影像分割。E-mail:zhwhong1022@163.com.责任作者：朱元骏（1978-），男，湖北十堰人，博士，副研究员，研究方向为土壤侵蚀。E-mail:zhuyuanjun@foxmail.com.基金项目：中国科学院重点部署资助项目（KFZD-SW-306-03）。