摘要
针对原木端面存在相交圆、阴影等导致原木轮廓难以获得完整的圆轮廓问题,提出了基于圆弧的原木轮廓识别方法.首先分析了圆弧在数字化图像中的性质,从各种边缘线条中区分出圆弧边缘;其次,定义验证模型,以判断识别结果是否为合格圆.结果表明,该方法可以从大量干扰的边缘中找到轮廓,并利用部分轮廓边缘识别原木端面.
To address difficulty in outline identification due to intersect circles and shadow on the cross section of bundled log, a method of outline extraction of logs cross section based on arc edges was proposed. At first, properties of arc edges in digital image were illustrated to identify real arc edge. Then, confirmation model was given to estimate whether the recognized circles were the out-lines of real log cross section. Validation results showed that the proposed method was able to eliminate interferences and effectively target the outline of logs.
出处
《福建农林大学学报（自然科学版）》
CSCD
北大核心
2016年第6期649-654,共6页
Journal of Fujian Agriculture and Forestry University:Natural Science Edition
基金
福建省自然科学基金资助项目(2014J01073)
福建省教育厅资助项目(JA12103)
国家自然科学基金资助项目(31300473
41401458)
国家林业局948资助项目(2013-4-65)
关键词
原木端面图像
轮廓识别
检尺自动化
验证
image of cross section of bundled logs
outline extraction
automatic scaling
validation
