期刊文献+

单目步进旋转式平台中的荔枝果图像预处理方法

Image pretreatment methods for litchi in single eye step-rotating platform
下载PDF
导出
摘要 针对荔枝果品质检测领域中传统人工方法费时、耗力和精度低等问题,构建了一套单目步进旋转式图像采集平台,研究了单颗荔枝果图像预处理过程中的相关算法,并对部分算法的效果进行了比较.结果表明:灰度线性变换法对于降低背景可见度的效果最好,对比度受限的自适应直方图均衡方法能有效提高目标的亮度与清晰度,线性平均滤波法的效果与Wiener法相当,但运行时间更短;改进型全局阈值法则能剔除由置物台造成的大部分阴影;形态学滤波算法可以将目标以内部全连通的形式完全提取出来;Canny算法能提取出荔枝果的完整外形轮廓.由此证明,在预处理流程中,对裁剪后的目标图像按顺序使用灰度线性变换、自适应直方图均衡、线性平均滤波、形态学滤波和Canny算法进行处理,可基本满足图像预处理的要求. Litchis usually needs to be classified into different quality tevels by detecting their qualities before being directly sold and reprocessed to be non-staple foodstuff, mainly depending on manual meth- ods currently, which brings about low accuracy rate, time consuming and high costs. A single eye step- rotating image collecting system, consisted of an electronic control subsystem and a mechanical platform, was constructed. The algorithms and processing procedures were studied and tested. According to the re- sults of these experiments : linear grey level transformation algorithm could efficaciously reduce the visibil- ity of background, contrast limited adaptive histogram balanced algorithm was able to enhance the bright- ness and articulation of targets, Linear mean filtering algorithm could decrease the noises in images, im- proved global threshold algorithm had the ability to eliminate shades caused by the rotation platform, mor- phology filtering algorithm could wipe off images' impurities and filling gaps, canny algorithm was able to extract the contour of litchis, which might meet the demands of image pretreatment.
出处 《仲恺农业工程学院学报》 CAS 2016年第2期50-55,共6页 Journal of Zhongkai University of Agriculture and Engineering
基金 广东省公益研究与能力建设专项(2014A020208139 2015A020214021 2015A020209175 2016B020202008) 广东省协同创新与平台环境建设专项(2016A040402044)资助项目
关键词 步进旋转式平台 荔枝果 品质检测 图像 预处理方法 step-rotating platform litchi quality detection image pretreatment methods
  • 相关文献

参考文献14

  • 1胡卓炎,余小林,赵雷,邓振权,黄辛,覃群明,蔡建兴,李云昌.荔枝龙眼主要加工产品生产现状[J].中国果业信息,2013,30(6):13-19. 被引量:19
  • 2陈厚彬,庄丽娟.关于促进我国荔枝产业发展的政策建议[J].中国热带农业,2011(1):14-15. 被引量:22
  • 3MAKOTO O,ATSUSHI M,KIICHIRO U,et al.Nondestructive measurement of sugar content in apples by millimeter-wave reflectometry[J].J Infrared Milli Terah Waves,2012,33(2):228-236.
  • 4GMEZ-SANCHIS J,LORENTE D,SORIA-OLIVAS E,et al.Development of a hyperspectral computer vision system based on two liquid crystal tunable filters for fruit inspection.application to detect citrus fruits decay[J].Food Bioproc Technol,2014,7(4):1047-1056.
  • 5CUBERO S,ALEIXOS N,ALBERT F,et al.Optimized computer vision system for automatic pre-grading of citrus fruit in the field using a mobile platform[J].Prec Agr,2014,15(1):80-94.
  • 6殷勇,陶凯,于慧春.基于机器视觉的苹果分级中特征参量选择方法[J].农业机械学报,2012,43(6):118-121. 被引量:18
  • 7杨丹,赵海滨,龙哲.MATLAB图像处理实例详解[M].北京:清华大学出版社,2013.
  • 8GONZALEZ R C,WOODS R E,EDDINS S L.Digital Image Processing Using MATLAB[M].2nd ed.New York:Gatesmark Publishing,2009:1-2,177-178.
  • 9PIZER S M,AMBURN E P,AUSTIN J D,et al.Adaptive histogram equalization and its variations[J].Comput Vision Graphics Image Pro,1987,39(3):355-368.
  • 10ZUIDERVELD K.Contrast limited adaptive histogram equalization:Graphic gems IV[M].San Diego:Academic Press Professional,1994:474-485.

二级参考文献37

共引文献85

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