<p>Only pairs of images used for deriving signatures (training) and for classification (with and without signature extension) from the same season were included.</p
Comparison with existing research in-terms of classification accuracy while using 15-scene image dat...
<p>Word recognition accuracy and classification accuracy on 2 and 6 locations, for horizontal and ve...
<p>Comparison of accuracy rate of different features extracted with classification algorithms.</p
<p><i>Note:</i> **Accuracy with feature selection is significant higher than that without feature se...
Comparison of using different data compositions of synthetic images for training the classifier and ...
(A) 30 images per category for training. (B) 40 images per category for training. (C) 50 images per ...
(a) Model without image enhancement and transfer learning; (b) Model without image enhancement, but ...
<p>The recognition rates (RR) of MR_2DLDA with and without pre-filtering when each image of every cl...
(A) 20 images per category for training. (B) 30 images per category for training. (C) 40 images per ...
<p>Summary of the labeling accuracy obtained for the training set, for all the discrimination proble...
<p>Training time and classification time of whole data set (4,037,099 pixels) using various classifi...
In this paper we present a case study of co-training to image classification. We consider two scene ...
<p>(A), Histograms of the NSSs in the training images. (B), Accuracy on classifying each pair of sce...
<p>If the number of colors used is 4, this means only the 4 most frequent colors from 7 extracted co...
Conventional supervised classification of satellite images uses a single multi-band image and coinci...
Comparison with existing research in-terms of classification accuracy while using 15-scene image dat...
<p>Word recognition accuracy and classification accuracy on 2 and 6 locations, for horizontal and ve...
<p>Comparison of accuracy rate of different features extracted with classification algorithms.</p
<p><i>Note:</i> **Accuracy with feature selection is significant higher than that without feature se...
Comparison of using different data compositions of synthetic images for training the classifier and ...
(A) 30 images per category for training. (B) 40 images per category for training. (C) 50 images per ...
(a) Model without image enhancement and transfer learning; (b) Model without image enhancement, but ...
<p>The recognition rates (RR) of MR_2DLDA with and without pre-filtering when each image of every cl...
(A) 20 images per category for training. (B) 30 images per category for training. (C) 40 images per ...
<p>Summary of the labeling accuracy obtained for the training set, for all the discrimination proble...
<p>Training time and classification time of whole data set (4,037,099 pixels) using various classifi...
In this paper we present a case study of co-training to image classification. We consider two scene ...
<p>(A), Histograms of the NSSs in the training images. (B), Accuracy on classifying each pair of sce...
<p>If the number of colors used is 4, this means only the 4 most frequent colors from 7 extracted co...
Conventional supervised classification of satellite images uses a single multi-band image and coinci...
Comparison with existing research in-terms of classification accuracy while using 15-scene image dat...
<p>Word recognition accuracy and classification accuracy on 2 and 6 locations, for horizontal and ve...
<p>Comparison of accuracy rate of different features extracted with classification algorithms.</p
DOI
Add to ORKG