期刊文献+

基于标记的极半径极值红枣形状识别方法 被引量:5

Identification of the Shape of Chinese Date Based on Labelling Method and Extremum of Polar Radius
下载PDF
导出
摘要 形状是分级的最重要参数之一,本文采用标记法对红枣形状进行了识别。通过图像预处理获取红枣二值图像,通过边界追踪获取目标边界笛卡尔坐标,并将其转化为极坐标,对目标图像进行缩放旋转使均值圆成为基线,切割的4部分边界曲线能完整表达。对边界曲线进行多项式拟合,获取极值点坐标,将其映射回被拟合曲线上,获取对应极值点坐标。若两极小极半径差值大于阈值,则红枣畸形;若两极大极半径附近区域极半径过渡平缓,判红枣为规整,否则为较规整。取53粒红枣进行检测,其中16粒畸形,17粒较规整,20粒规整。检测结果表明:畸形枣识别准确率达100%,较规整枣的识别准确率94%,规整枣识别准确率95%,可基本满足红枣分级系统精度的要求。 Shape is a basic parameters of classification, this paper has a identification on shape of Chinese date based on the mark method.Have a image preprocessing to get binary image of Chinese date;Have a boundary tracking to get car-tesian coordinates of target boundary; Transform them into polar coordinates; Scale and rotate the target image to make mean round become a base line and cut the boundary curve to four parts by it.Have a polynomial fitting on the boundary curves in order to get extreme value point coordinates;Mapping them to the fitted curve to obtain corresponding points.If the difference value of two local minimum polar radius is bigger than the threshold value , the Chinese date is deformity;If the points nearby two local minimum polar radius transit gently, the Chinese date is considered as regular shape, else relatively regular shape.Do experiments with 53 Chinese dates including 16 abnormal shape dates and 17 relatively regu-lar shape dates and 20 normal dates.The result show that the recognition rate of deformity jujube is 100% and that of relatively normal jujube is 94 and that of normal jujube is 95 , which basically meet the accuracy requirement of the red jujube grading system.
作者 肖爱玲 潘斌
出处 《农机化研究》 北大核心 2015年第7期61-65,共5页 Journal of Agricultural Mechanization Research
基金 国家级大学生创新训练计划项目(2013107570015)
关键词 红枣 标记 极半径 形状检测 chinese date labelling method polar radius shape detection
  • 相关文献

参考文献14

  • 1DunhamJ G. Optimum uniform piecewise linear approximation of planar curves [ J ]. IEEE Trans. Pattern Anal. Ma- chine Intell, 1986-PAMI ( 8 ) :67-75.
  • 2Zayas I, Converse H Steele J. Discrimination of whole from Broken corn kernels with image analysis [ J ]. Trans of the ASAE, 1990,5 ( 4 ) :445-445.
  • 3Vorren V D,J G G Polder, W A M Van Der Heijden. Iden- tification of mushroons cultivars using image analysis [ J ]. Transactions of the ASAE, 1992,35 ( 1 ) :347-350.
  • 4Liao K. , M. R. Paulsen,J. F. Reid,et al. Corn kernel break- age classification by machine vision using a neural network classifier[ J ]. Transactions of ASAE, 1993,36 ( 6 ) : 1949 - 53.
  • 5Heinemann P H,Z A Varghese,C T Marrow, et al. Machine vision inspection of golden delicious apples [ J ]. Applied En- gineering in Agriculture, 1995,11 (6) :901-906.
  • 6Tao Y, C T Morrow, P H Heinemann, et al. Fourier-based separation technique for shape-grading of potatoes using ma- chine vision[ J ]. Transactions of the ASAE, 1995,38 (3) : 949 -957.
  • 7Ni B,M R Paulsen,J F Reid. Corn kernel crown shape iden- tification using image processing [ J ]. Trans of the ASAE, 1997,40(3 ) :833-838.
  • 8Luo X, D D Jayas, N r Bulley. Identification of damaged ker- nels in wheat using a color machine vision system [ C ]//In: 1997 ASAE Annual International Meeting. Minnesota: ADAE. 1997.
  • 9Paulus I, E Schrevens. Shape characterization of new apple cultivars by Fourier expansion of digitized images [ J ]. Jour- nal of agricultural engineering research, 1999,72: 113- 118.
  • 10Shouehe S p, Rastogi R, Bhagwat S G, et al. Shape analysis of grains of Indian wheat varieties [ J ]. Computers and Elec- tronics in Agriculture,2001,33:55-76.

二级参考文献18

共引文献46

同被引文献68

引证文献5

二级引证文献28

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部