Table 7 Scene classification rate (%) in Scene 15
From: DTCTH: a discriminative local pattern descriptor for image classification
Techniques | Accuracy |
|---|---|
Places-CNN [71] | 90.19 |
ImageNet-CNN [71] | 84.23 |
Hybride-CNN [71] | 91.59 |
SIFT (SPM + pLSA) SVM, 2006 [5] | 81.40±0.50 |
Dense color SIFT (pLSA) SVM, 2008 [34] | 72.70 |
Dense color SIFT (SP-pLSA) SVM, 2008 [34] | 83.70 |
Dense SIFT (KC) SVM with HI, 2008 [12] | 77.10 |
Dense SIFT (LSPM + MP) LSVM, | |
2009 [60] | 65.32±1.02 |
Dense SIFT (ScSPM + MP) LSVM, | |
2009 [60] | 80.28±0.93 |
Dense SIFT (Sparse Code) LSVM, | |
2010 [96] | 84.10±0.50 |
Dense SIFT (LLC + MP) LSVM, | |
79.81±0.35 | |
Dense SIFT (LSA + MMP) LSVM, | |
2011 [68] | 82.70±0.39 |
SIFT (BOVW + SPCK++) SVM, | |
2011 [57] | 82.51±0.43 |
Dense SIFT (LDC + LLC/LSA + MP) | |
LSVM, 2013 [64] | 82.50±0.47 |
Dense SIFT (LCSR + MP) LSVM, | |
2012 [61] | 82.67±0.51 |
Object bank + LSVM, 2010 [55] | 80.90 |
Dense SIFT + I2CDML, 2010 [58] | 77.00±0.60 |
Dense SIFT (SPM) I2CDML, 2010 [58] | 81.20±0.52 |
72.30±0.93 | |
PRICoLBP + SVM with χ 2, 2014 [8] | 82.04 |
Dense SIFT (BoF) SVM with HI, | Â |
2014 [93] | 82.06 |
LGP (SPM) LSVM, 2013 | 78.22±0.56 |
OC-LBP (BoF) LSVM, 2013 | 77.22±0.40 |
LAID (SPM) LSVM, 2013 | 81.18±0.60 |
CLBP_S/M/C (SPM) LSVM, 2010 | 76.47±0.15 |
LTP (SPM) LSVM, 2010 | 80.25±0.31 |
GIST + LSVM, 2001 | 55.55±0.67 |
CENTRIST (SPM) LSVM, 2011 | 81.45±0.23 |
Proposed (DTCTH + LSVM) | 82.66±0.50 |
Proposed (DTCTH + HI) | 83.63±0.21 |