Fig. 1. Bad pixel 3D image
Fig. 2. Gray value curves of bad pixel and neighboring normal pixels
Fig. 3. Gray gradient value curves of bad pixel and neighboring normal pixels
Fig. 4. The infrared image of star target
Fig. 5. 3D image of bad pixel in infrared image
Fig. 6. The detection window of bad pixels
Fig. 7. The compensation window of bad pixels
Fig. 8. Background 1
Fig. 9. Background 2
Fig. 10. The detection results based on gray value of different background
Fig. 11. The detection results based on gray gradient value of different background
Fig. 12. The light bad pixels in both background and star target
Fig. 13. The detection results of bad pixels
Fig. 14. The neighboring bad pxiels
Fig. 15. The detection results of bad pixels
Fig. 16. The compensation results of bad pixels
Fig. 17. The multiple bad pixels in star image
Fig. 18. The detection results of bad pixels
Fig. 19. The compensation results of bad pixels
Type of bad pixel | Centroid coordinates | X coordinate error | Y coordinate error |
---|
None | (15.065 6,14.878 6) | / | / | Light bad pixel in background | (15.065 7,14.878 3) | 0.001% | 0.002% | Dark bad pixel in background | (15.065 5,14.878 7) | 0.001% | 0.001% | Light bad pixel in star | (14.845 8,15.109 8) | 1.459% | 1.554% | Dark bad pixel in star | (15.092 5,14.850 4) | 0.179% | 0.190% |
|
Table 1. The influence of bad pixels on centroid coordinate extraction of star target
Method | Number of bad pixels | Number of detected bad pixels | Accuracy |
---|
The compared method | 5 | 2 | 40% | Our method | 5 | 5 | 100% |
|
Table 2. The detection accuracy of bad pixels
Method | Centroid coordinates | X coordinate error | Y coordinate error |
---|
None bad pixel | (15.0656,14.8786) | / | / | Baseline | (14.8451,15.1103) | 1.464% | 1.557% | The compared method | (15.0013,14.9460) | 0.427% | 0.453% | Our method | (15.0409,14.9046) | 0.164% | 0.175% |
|
Table 3. The result of extracting star's centroid coordinates