2017 IEEE 7th International Advance Computing Conference (IACC) 2017
DOI: 10.1109/iacc.2017.0132
|View full text |Cite
|
Sign up to set email alerts
|

A Feature Subset Based Decision Fusion Approach for Scene Classification Using Color, Spectral, and Texture Statistics

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
4
0

Year Published

2018
2018
2024
2024

Publication Types

Select...
4
2

Relationship

0
6

Authors

Journals

citations
Cited by 6 publications
(4 citation statements)
references
References 22 publications
0
4
0
Order By: Relevance
“…On the other hand, soft decision schemes, such as linear opinion pool and logarithmic opinion pool, use the posterior probabilities of each classifier (without relaying on class label of each of them) to generate the central decision (final ensemble result) [67]. In a recent study, a hierarchical decision fusion scheme is proposed to classify an outdoor scene [68]. This scheme consists of two stages: the first uses Gabor feature extraction to extract the scenes into artificial or natural images, while the second stage fuses the decisions of gradient binary patterns, grey level co-occurrence matrix, and color features for further classification (i.e., to identify forest, coast, mountain, and open country scenes).…”
Section: B Ensemble Learningmentioning
confidence: 99%
“…On the other hand, soft decision schemes, such as linear opinion pool and logarithmic opinion pool, use the posterior probabilities of each classifier (without relaying on class label of each of them) to generate the central decision (final ensemble result) [67]. In a recent study, a hierarchical decision fusion scheme is proposed to classify an outdoor scene [68]. This scheme consists of two stages: the first uses Gabor feature extraction to extract the scenes into artificial or natural images, while the second stage fuses the decisions of gradient binary patterns, grey level co-occurrence matrix, and color features for further classification (i.e., to identify forest, coast, mountain, and open country scenes).…”
Section: B Ensemble Learningmentioning
confidence: 99%
“…In [6] Anish et.al, proposed multi stage decision based fusion approach for texture classification using global and local features with SVM classifier.…”
Section: Related Workmentioning
confidence: 99%
“…where ŷ, y, r represent the fused class label, class label, and the total number of classifiers, respectively. Several feature and decision fusion methodologies have been proposed in the literature to improve the classifica tion accuracy of the recognition system locally and globally [21,44]. For example, a softmax regressionbased tech nique, called marginalized kernel, has been proposed to fuse spectral, texture, and localself similarity descriptors of the QuickBird imagery linearly.…”
Section: Feature and Decision Fusionmentioning
confidence: 99%
“…Each fea ture vector is encoded using a bagofvisualwords method to unify all feature spaces (no normalisation is required) [21]. On the other hand, the decisions of Gabor features, gradient binary patterns, grey level cooccurrence matrix, and colour features are merged hierarchically to classify outdoor scenes such as forest, coast, mountain, and open country scenes [44].…”
Section: Feature and Decision Fusionmentioning
confidence: 99%