向量数据描述在烟叶异物检测中的应用.docVIP

支持向量数据描述在烟叶异物检测中的应用 摘 要:针对烟叶异物检测中很难全面收集异物样本数据的问题,提出一种基于支持向量数据描述方法(svdd)的烟叶异物检测方法。该方法只需要烟叶样本数据,就可建立单值分类器。首先,提取烟叶与几种典型异物的rgb分量与hsv分量;然后,选取烟叶的hv分量作为特征向量,训练svdd分类器,实现烟叶异物的分类识别;最后,通过接受者操作特性(roc)曲线对比了svdd与其他3种方法的分类效果。实验结果表明,采用hv分量降低了数据维数,提高了计算效率;svdd方法具有很好的分类效果和计算效率,能很好地区分烟叶与异物。 关键词:支持向量数据描述;异物检测;烟叶样本;hv分量;分类识别  application of support vector data description to detection of foreign bodies in tobacco  huang shijian1,2* ( 1.school of physics and electron engineering, yangtze normal university, chongqing 408100, china; 2.key laboratory of optoelectronic technology and systems (chongqing university), ministry of education, chongqing 400044, china ) abstract: it is difficult to fully collect foreign body sample in detecting foreign bodies from tobacco. a detection method based on support vector data description (svdd) was proposed. thus a oneclass classifier can be developed by using tobacco samples soly. rgb and hsv of tobacco and several typical foreign bodies were firstly extracted; then the hv component was used as eigenvector. a developed svdd classifier was applied to distinguish foreign bodies from tobacco by inputting the hv eigenvector. finally through the receiver operating characteristic (roc) curve, the svdd classifier was compared with three other methods in classification effect. the experimental results show that by adopting feature extraction with hv component, the data dimension was reduced and a higher computation efficiency was achieved. the svdd classifier has a stronger classification ability and higher efficiency, which could distinguish foreign bodies from tobacco better. key words: support vector data description (svdd); foreign body detection; tobacco sample; hv component; classification  0 引言 国内大多数烟草生产线上都采用金属探测仪检测和剔除金属异物,并辅之以人工剔除其他异物的方法[1-2],这些方法的异物剔除率受人为因素影响较大。因此研究烟叶异物自动检测系统极为重要,自动检测的关键技术多采用机器视觉技术与模式识别技术相结合[3],其中分类方法的选择是否恰当,直接影响最后的检测结果。但近年来国内外针对烟叶异物的分类技术却鲜有报道,而常用的可以借鉴的分类方法有贝叶斯分类算法、支持向量机等传统方法和支持向量数据描述[4-6](support vector d