基于外部特征的玉米品种计算机识别系统

Recognition System for Corn Species by Exterior Parameters

用基于外部特征参数的计算机自动识别方法来替代目前广泛使用的品种形态鉴定法 ,以提高玉米品种识别效率和精度 ;并利用数理统计和模糊数学知识建立了反映玉米品种间特征参数相似程度的特征贴近度、品种贴近度计算公式和品种判别准则 ;基于 3类 11个外部特征参数构造出成本低、使用简便的玉米品种计算机识别系统。在 5个品种 5 0次实验检测中 ,正确率达 96 % ,品种正确识别率达 88%。

An idea of using computer to identify corn species by detecting corn external features was put forward to improve the identification precision and the efficiency of a morphological feature identification method widely used at present. The species identification principle and the computing formula of feature close degree (the approximate degree between corn species feature parameters), and species close degree were developed using mathematical statistics and fuzzy mathematics. A low-cost, easily operated corn species identification system was constructed based on the parameters from 11 external features of 3 categories. The detection experimental results from total 50 runs for 5 corn species showed that the conclusion accuracy of the system was up to 96%, and the species identification accuracy was up to 88%.