Table 1
Values of the Parameters that were Used in the Tracking Stage (Subsection 3B); These Values were Used for Both the Degraded and Restored Video Sequences
Parameter Name | Parameter Value |
---|
N
| 15 (frames) |
THRESH_TURB | 0.1 (pixels) |
THRESH_CM | 4 (pixels) |
THRESH_SIZE | 1.5 |
Table 2
Average Precision Values [Eq. (8)], Averaged for All Sequence Frames, Obtained for the Human Target [Whose Ground-Truth is Shown in Fig. 1(b)] in Different Imaging Conditions and Video Types (Sample Frames of These Videos Are Shown in Fig. 2) Using the Median Filter for Background Estimation (Subsection 3A) and Using the Nonparametric Model [23, 24] for Background Estimation; The Average Recall Value [Eq. (7)] in All Cases Was Set to the Same Value (0.99 in Our Implementation)
Video Type Imaging Conditions | Degraded | Restored (with Estimated PSF) | Restored (with True PSF) |
---|
Using the median filter |
Low noise and high blur | 0.33 | 0.52 | 0.54 |
High noise and high blur | 0.32 | 0.39 | 0.40 |
Low noise and low blur | 0.63 | 0.92 | 0.97 |
High noise and low blur | 0.61 | 0.63 | 0.67 |
Using the nonparametric model |
Low noise and high blur | 0.38 | 0.56 | 0.57 |
High noise and high blur | 0.35 | 0.41 | 0.42 |
Low noise and low blur | 0.70 | 0.96 | 0.99 |
High noise and low blur | 0.68 | 0.69 | 0.73 |
Table 3
Average Precision Values, Averaged for All Sequence Frames, Obtained for the Car Target [Whose Ground-Truth is Shown in Fig. 1(c)] in Different Imaging Conditions and Video Types Using the Median Filter and Nonparametric Model for Background Estimation. The average recall value was set to 0.99.
Video Type Imaging Conditions | Degraded | Restored (with Estimated PSF) | Restored (with True PSF) |
---|
Using the median filter |
Low noise and high blur | 0.76 | 0.86 | 0.86 |
High noise and high blur | 0.73 | 0.75 | 0.76 |
Low noise and low blur | 0.96 | 0.99 | 1.0 |
High noise and low blur | 0.94 | 0.95 | 0.96 |
Using the nonparametric model |
Low noise and high blur | 0.75 | 0.86 | 0.87 |
High noise and high blur | 0.72 | 0.74 | 0.75 |
Low noise and low blur | 0.96 | 0.99 | 1.0 |
High noise and low blur | 0.95 | 0.95 | 0.95 |
Table 4
Description of the Five Moving Objects that Appear in the Recorded Video Sequences; These Objects are Numbered and Marked by Circles in the Sample Frames Shown in Fig. 3
Moving Object Number | Moving Object Description |
---|
1 | Bicycle rider |
2 | Person with a pole |
3 | Dog |
4 | Walking man |
5 | Man on a roof |
Table 5
Dispersedness (perimeter2∕area) Values for Each of the Five Moving Objects that are Numbered and Marked by Circles in the Sample Frames Shown in Fig. 3; It Can be Noted that the Restoration Improved the Dispersedness Values by an Average of ∼28%
Moving Object Number | Dispersedness-Blurred Frame | Dispersedness-Restored Frame |
---|
1 | 40.76 | 55.20 |
2 | 41.32 | 45.73 |
3 | 32.31 | 32.38 |
4 | 32.37 | 47.26 |
5 | 19.13 | 28.06 |