《Table 3 Results of the three classification models for detection of potential fruit》

  提示：宽带有限、当前游客访问压缩模式

本系列图表出处文件名：随高清版一同展现

《Feature extraction of hyperspectral images for detecting immature green citrus fruit》

1. 获取 高清版本忘记账户？点击这里登录

1. 下载图表忘记账户？点击这里登录

The aim of a classifier is to minimize the FP and FN or to maximize the TP，TN，accuracy，sensitivity and specificity.Table 3 illustrates the performance of the three classification models on the test data set.It could be concluded that the SVM classifier had the best classification ability.Among 27 fruit samples and 50 leaves samples，eight fruit were misclassified as leaves and three leaves were misclassified as fruit.The accuracy，sensitivity and specificity score were 0.86，0.94 and 0.70，respectively.