Table 4.
Comparative WBC detection among methods that considers the complete data set of 30 images corrupted by different levels of Gaussian noise.
| Noise level | Method | Leukocytes detected |
Missing | False alarms | DR | FAR |
|---|---|---|---|---|---|---|
|
σ = 5 Gaussian noise 426 Leukocytes |
BSV | 172 | 254 | 77 | 40.37% | 18.07% |
| IO | 309 | 117 | 71 | 72.53% | 16.67% | |
| Wang | 301 | 125 | 65 | 70.66% | 15.26% | |
| BGA | 345 | 81 | 61 | 80.98% | 14.32% | |
| EMO | 397 | 29 | 21 | 93.19% | 4.93% | |
|
| ||||||
|
σ = 10 Gaussian noise 426 Leukocytes |
BSV | 143 | 283 | 106 | 33.57% | 24.88% |
| IO | 281 | 145 | 89 | 65.96% | 20.89% | |
| Wang | 264 | 162 | 102 | 61.97% | 23.94% | |
| BGA | 308 | 118 | 85 | 72.30% | 19.95% | |
| EMO | 380 | 46 | 32 | 89.20% | 7.51% | |