KSRCKernel Source
KSRCKent Science Resource Centre (UK)
KSRCKashima Space Research Center (Japan)
KSRCKawasaki Sportsbike Riders Club (Australian Kawasaki web forum)
KSRCKentucky Society for Respiratory Care (Lexington, KY)
KSRCKeyStone Research Corporation (Erie, PA)
References in periodicals archive ?
Similar to the KSRC, suppose that [phi](y) is the representation of query pixel in the high-dimensional feature space; the SRBBH model becomes
In this section, the classification performance of KSRBBHC is evaluated and compared to the other four classifiers (SVMC, SRC, KSRC, and SRBBHC), and RBF kernel function [kappa]([x.sub.i], [x.sub.j]) = exp (-[[parallel][x.sub.i] - [x.sub.j][parallel].sup.2]/[[sigma].sup.2]) is used for KSRC and KSRBBHC.
We compute the ARRs and ORRs of KSRC and KSRBBHC for ROI-I with varying kernel parameter a and sparsity level [K.sub.0], as shown in Figure 1.
The ARRs and ORRs of KSRC and KSRBBHC with varying a and [K.sub.0] for ROI-II are shown in Figure 2.
For the ROI-II, the classification performances of SRC, KSRC, and SRBBHC increase with [K.sub.0] when the parameter is smaller than 6; then the performances of SRC and KSRC keep increasing smoothly with [K.sub.0], whereas the performance of SRBBHC reduces gradually.
In detail, for ROI-I, because of the mixture of desired target and background, the SVMC, SRC, and KSRC mistake desired target to background to a great extent.
Caption: Figure 1: Effect of sparsity level [K.sub.0] and RBF kernel parameter [sigma] on performance of two kernel- based classifiers: (a) KSRC; (b) KSRBBHC for ROI-I.
We compare the performance of the proposed MKSRC with the state-of-the-art classifiers, such as SVM [41], SRC [22], KSRC (Polynomial) [44], KSRC (Gaussian) [44].
We can see that MKSRC algorithm retains higher performance than SRC and KSRC. In dimension 300, the recognition rate of MKSRC is 69.32% which is 4.5% higher than SVM.
From the experiment without occlusion, we can see the proposed MKSRC method not only outperforms the SRC, but also outperforms the KSRC. Experiments results demonstrate that kernel information helps to improve the recognition rate.
On the high-dimensional data such as face images, KSRC algorithm has got better performance than SRC, but KSRC algorithm does not make full use of kernel information.
Accuracy on FERET face database SVM SRC KSRC (Polynomial) Dimensions(d= 50) Accuracy 57.5% 54.3% 57.3% Parameter d=2 Dimensions(d=200) Accuracy 59.2% 66% 68.2% Parameter d=2 Dimensions(d=300) Accuracy 64.8% 65.5% 67.1% Parameter d=2 KSRC MKSRC (Gaussian) Dimensions(d= 50) Accuracy 56.4% 59.8% Parameter t=2 d=2 t=2 Dimensions(d=200) Accuracy 67.1% 70.3% Parameter t=2 d=2 t=2 Dimensions(d=300) Accuracy 65.6% 69.32% Parameter t=2 d=2 t=2 Table 2.