Research Article
A Real-Time Framework for Human Face Detection and Recognition in CCTV Images
| | No. of features | Training data | Numerical methods | k = 1 | k = 2 | k = 3 | k = 4 | k = 5 |
| | 5 | 90 | Euclidean | 89.0115% | 89.7889% | 79.0841% | 76.5861% | 75.278% | | Manhattan | 94.7623% | 90.0457% | 88.6113% | 86.9326% | 86.0086% | | 80 | Euclidean | 87.8664% | 80.2338% | 77.7842% | 75.2137% | 73.5403% | | Manhattan | 93.7989% | 89.0839% | 87.6401% | 85.9456% | 84.8975% |
| | 10 | 90 | Euclidean | 88.3589% | 79.7717% | 77.8163% | 76.1214% | 75.0927% | | Manhattan | 93.7989% | 89.4494% | 88.4072% | 87.1582% | 77.0927% | | 80 | Euclidean | 86.8185% | 77.905% | 75.9567% | 74.377% | 73.4407% | | Manhattan | 93.6811% | 88.3288% | 87.335% | 86.0392% | 85.3475% |
| | 15 | 90 | Euclidean | 86.383% | 76.6452% | 74.7327% | 73.293% | 72.5651% | | Manhattan | 93.9484% | 88.2299% | 87.3221% | 86.1644% | 85.618% | | 80 | Euclidean | 84.5773% | 74.3589% | 72.5164% | 71.1934% | 70.4926% | | Manhattan | 92.8172% | 86.6614% | 85.9425% | 84.7646% | 84.1484% |
|
|
