单叶快速定级等级错误去向量化分析  

Quantitative Analysis of Lamina Tobacco Fast Grading Error Rate

作　　者：王改丽[1] 甄焕菊[1] 于建军[2] 郑宪滨[1] 魏春阳[1] 李悦[2] WANG Gai-li ZHEN Huan-ju YU Jian-jun et al(Staff Development Institute of China National Tobacco Corporation, Zhengzhou, Henan 450008 College of Tobacco Science, Henan Agricultural University, Zhengzhou, Henan 450002)

机构地区：[1]中国烟草总公司职工进修学院,河南郑州450008 [2]河南农业大学烟草学院,河南郑州450002

出　　处：《安徽农业科学》2017年第6期61-64,共4页Journal of Anhui Agricultural Sciences

基　　金：中国烟草总公司职工进修学院课题"烟叶分级技术的隐性知识显性化研究"

摘　　要：[目的]突破烟叶分级技术能力提升的瓶颈。[方法]采用统计分析法,对单叶快速定级的等级错误率及错误去向进行量化分析。[结果]等级错误率以B1L最高,达到79.6%,其次为等级X1F(66.2%)、B1F(63.9%)、X1L(62.1%)和C4L(60.2%)。等级X2F、X2L,B2L和C4L的错误去向分别为C3F(8.1%)、C3L(9.3%)、C2F(10.8%)和X2L(15.9%),属于部位易错等级;等级X3L和X4L的错误去向分别为X3F(12.1%)和X4F(19.6%),属于颜色易错等级。其中,等级B3F与B4F、B3L与B4L、C2F与C3F、C2L与C3L、X3F与X4F、B2K与B3K、GY1与GY2这7组等级互为易错,属于易混淆等级。[结论]该研究可为找准单叶快速定级疑难点提供数据支撑。[ Objective ] To break through the bottleneck of tobacco grading technology improvement. [ Method ] Using statistical method,grade error rate and wrong orientation in lamina tobacco fast grading test were quantitatively analyzed. [ Result ] Grade B1L had the highest error rate, which at the level of 79.6% ,followed by grade X1F （66.2%） ,B1F （63.9%） ,X1L （62. 1% ） ,and C4L （60.2%）. Grade X2F,X2L,B2L and CAL respectively mistake to C3F （8.1%） ,C3L （9.3%） ,C2F （ 10.8% ） and X2L （ 15.9% ） ,which belonged to the position error-prone grades; Grade X3L and X4L respectively error to X3F （12.1%） and X4F （19.6%） ,which belonged to the color error-prone grades. Cos group B3F and IMF,B3L and IML,C2F and C3F,C2L and C3L,X3F and X4F,B2K and B3K,GY1 and GY2 respectively error to each other,which belonged to the easy confusing grades. [ Conclusion] The study can provide data support for getting the difficult points in lamina tobacco fast grading test.

关 键 词：烤烟分级 等级错误率 等级错误去向 

分 类 号：TS442[农业科学—烟草工业]