Identification and Classification of Lungs Focal Opacity Using CNN Segmentation and Optimal Feature Selection

Table 6

LBP-HOG-Size features-based classification results.

Methods	Prediction speed (obs/sec)	Training time (sec)	Accuracy (%)
Bagged ensemble	220	7.71	88.8
Subspace discriminant	160	9.043	91.4
Subspace KNN	120	14.98	88.8
RUSBoost	360	16.93	89.5
Fine tree	710	2.16	80.9
Medium tree	790	1.57	80.9
Coarse tree	1000	3.03	86.2
Linear SVM	560	1.161	90.1
Quadratic SVM	620	1.618	97.8
Cubic SVM	550	3.79	89.5
Fine Gaussian SVM	760	3.59	80.9
Medium Gaussian SVM	750	4.03	85.5
Fine KNN	680	2.4	81.6
Medium KNN	570	1.73	74.3
Cubic KNN	630	3.63	80.9
Cosine KNN	340	3.73	77.6
Weighted KNN	630	3.27	84.2