《Table 7 Classification accuracies of the proposed method for the two datasets》
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
《D-SS Frame: deep spectral-spatial feature extraction and fusion for classification of panchromatic and multispectral images》
The ground truth map of‘Washington DC’and‘Stockholm’data[14]were constructed.In each labeled scene,the black area represents the unlabeled area.Different values of the SSAE and SAE parameters were experimented,and Tables 3,4,5,6 list the classification accuracy of(SSAE+SVM)and(SAE+SVM)for MS and PAN images of Washington DC and Stockholm data respectively,versus the number of neurons in a hidden layer.
图表编号 | XD0020606900 严禁用于非法目的 |
---|---|
绘制时间 | 2018.12.01 |
作者 | Teffahi Hanane、Yao Hongxun |
绘制单位 | School of Computer Science and Technology,Harbin Institute of Technology、Algerian Space Agency、School of Computer Science and Technology,Harbin Institute of Technology |
更多格式 | 高清、无水印(增值服务) |