《Table 3 Recognition accuracy w ith different feature dimensions》
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
《Individual Dairy Cattle Recognition Based on Deep Convolutional Neural Network》
To verify the effect of feature dimension tow ards recognition results,different feature dimensions in full connection layer are chosen for experiments.The feature w ill not be described w ell if the dimension is small just as the results of choosing 1 024,2 048,and 3 072 show n in Table 3.When the feature dimension is 4 096,the recognition accuracy is good.Continuing to increase the dimension,the training sample recognition rate is better but the training sample result is not ideal because too big feature dimension makes over-fitting phenomenon[38].
图表编号 | XD0014946800 严禁用于非法目的 |
---|---|
绘制时间 | 2018.04.30 |
作者 | 张满囤、单新媛、于津苏、郭迎春、李睿文、徐明权 |
绘制单位 | School of Computer Science and Engineering,Hebei University of Technology、Hebei Province Key Laboratory of Big Data Calculation、School of Computer Science and Engineering,Hebei University of Technology、Hebei Province Key Laboratory of Big Data Calculation |
更多格式 | 高清、无水印(增值服务) |
查看“Table 3 Recognition accuracy w ith different feature dimensions”的人还看了
- Table 4 Recognition accuracy and confusion matrix of different fusion-strategy models表4不同融合策略模型的识别准确率和混淆率