基于机器视觉的马铃薯图像特征参数获取方法.PDFVIP

Electrical Engineering andAutomation December 2016, Volume 5, Issue 2, PP.42-49 Methodfor ObtainingCharacteristicParameters ofPotatoImageBasedon MachineVision 1 1†, 2, 1, 1 1 ZouzouLi ,Hongjun Wang JuntaoXiong Jianmeng Deng Yuanhong Li , Weiliang Zhou 1.CollegeofEngineering, South ChinaAgriculturalUniversity,Guangzhou,510642,China 2.CollegeofInformatics, South ChinaAgriculturalUniversity,Guangzhou 510642,China †Email:xtwhj@ Abstract Potato is the fourth most worldwide crop in the world after wheat, corn and rice. Classification of potatoes is thought to produce greater economic benefits. The current grading methods mainly relies on artificial methods which is not stable and expensive. In recent years,machinevision technologieshavebeen widely used in the field of agriculture, especially achieved great resultsin the field ofagriculturalproductsclassification.Machinevision couldclassify differentvarietiesofagriculturalproductsbased on color, volume,quality and other characteristics.For the same variety ofpotatoes,the shapetype andweight range couldbegradedby the external dimensions. For this purpose, a set of machine vision system was designed, including CCD camera, image acquisition card, computer system, LED light source,two mirrorsplaced intoV type, as well as MATLAB software for image processing and Unscramble software for data analysis.Photoswereprocessed intobinary imageswith MATLAB software.Each number ofpixels oftarget area can be counted asits area.Theminimum external rectangle of each target area was obtained for calculating itslength and width. The area, length and width were regarded as the image characteristic parameters of the target area. A binary image con