1 / 19

Presented By: Anthony Karloff

Demosaicing with Improved Edge Direction Detection. Presented By: Anthony Karloff. Overview. Demosaicing Background Basics and Challenges Advanced Methods (State of the Art) Color Channel Reconstruction Conclusion. Demosaicing Background. Background. Why Image Reconstruction?. Basics.

laura-myers
Download Presentation

Presented By: Anthony Karloff

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Demosaicing with Improved Edge Direction Detection Presented By: Anthony Karloff

  2. Overview • Demosaicing Background • Basics and Challenges • Advanced Methods (State of the Art) • Color Channel Reconstruction • Conclusion

  3. Demosaicing Background Background Why Image Reconstruction? Basics • Incomplete color planes from CCD sensors. Advanced Methods Channel Reconstruction Conclusion Color Filter Array (CFA) on image sensor.

  4. Basics and Challenges Background Color Plane Interpolation Basics • Must Interpolate color planes to re-create image. Advanced Methods Channel Reconstruction Conclusion Red Channel Interpolation

  5. Basics and Challenges Background Color Plane Interpolation Methods Basics • Pixel Averaging • - lose image resolution R1 G1 Advanced Methods P1 G2 B1 Channel Reconstruction Conclusion R1 G1 P1 • Nearest Neighbor • - Poorest Quality G2 B1 G1 B1 G2 • Bilinear/Spline • - Color artifacts at edges R1 G3 R2 P1 G4 B2 G5

  6. Basics and Challenges Background Color Artifacts Basics • Problem with most simple interpolation algorithms is the presence of color artifacts. Advanced Methods Channel Reconstruction Conclusion Original Image Bilinear Interpolation of Bayer Image

  7. P1 P1 P2 P2 P3 P3 P4 P4 P5 P5 P6 P6 P7 P7 P8 P8 P9 P9 P10 P10 P11 P11 P12 P12 P13 P13 P14 P14 P15 P15 P16 P16 P17 P17 P18 P18 P19 P19 P20 P20 P21 P21 P22 P22 P23 P23 P24 P24 P25 P25 Basics and Challenges Background Color Artifacts Basics • Due to interpolation across edges. Advanced Methods Channel Reconstruction Artifacts Conclusion Bilinear Interpolation Dark to Light Edge over Bayer Pattern Resulting Edge after Interpolation

  8. Advanced Methods Background Advanced Techniques for Color Plane Interpolation Basics Advanced Methods • Use color plane gradients Channel Reconstruction • Group pixels of similar objects Conclusion • Interpolate along edges (not across) • Interpolate green color plane first • Interpolate image more than one iteration (refinement)

  9. P1 P2 P3 P4 P5 P6 P7 P8 P9 Advanced Methods Background Using Gradients for Image Reconstruction Basics Advanced Methods • Better estimation of color plane behavior. Channel Reconstruction Conclusion Bayer Pattern for Green Centered Pixel 1. GRADIENTS • Notice that the differences are always from the same color plane.

  10. P1 P2 P3 P4 P5 P6 P7 P8 P9 Advanced Methods Background Kimmels ‘E’ Function for Pixel Grouping Basics Advanced Methods • Associates colors of the same object. Channel Reconstruction Ie. If P5 and Pi are part of the same object, E will be close to unity. Conclusion Bayer Pattern for Green Centered Pixel • There are eight Ei values for each pixel. One for each neighbor. 1. GRADIENTS 2. GROUPING

  11. P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22 P23 P24 P25 Advanced Methods Background Using Edge Detection (Wide) Basics • Interpolation is best performed in the same direction as an edge. Advanced Methods Channel Reconstruction Edge detection of radius 3 Conclusion Bayer Pattern for Red Centered Pixel 1. GRADIENTS 2. GROUPING 3. EDGE DETECT 3 pxls

  12. P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22 P23 P24 P25 Advanced Methods Background Narrow Edge Detection Basics Advanced Methods • Uses narrow edge detection to improve edges by looking between color planes. Channel Reconstruction Edge detection of radius 2 Conclusion Bayer Pattern for Red Centered Pixel 1. GRADIENTS 2. GROUPING 3. EDGE DETECT 3 pxls 4. EDGE DETECT 2 pxls

  13. P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22 P23 P24 P25 Advanced Methods Background Local Inter-Channel Correlation Basics • Compare average color differences in a 5x5 region to determine whether the Red or Blue channel is more closely related to the green. Advanced Methods Channel Reconstruction Conclusion 1. GRADIENTS 2. GROUPING 3. EDGE DETECT 3 pxls 4. EDGE DETECT 2 pxls Bayer Pattern for Red Centered Pixel 5. COLOR CORRELATION

  14. Advanced Methods Background Improved Edge Detector Basics • Now we can complete the edge detector Advanced Methods 4 5 3 Channel Reconstruction Conclusion 1. GRADIENTS 2. GROUPING 3. EDGE DETECT 3 pxls 4. EDGE DETECT 2 pxls 6. IMPROVED EDGE DETECTION 5. COLOR CORRELATION

  15. Channel Reconstruction Background Channel Reconstruction Overview Basics • For each pixel we now have: Advanced Methods • Approximate the red and blue channels using Bilinear Interpolation. Channel Reconstruction Conclusion • Reconstruct the green channel using edge detectors and the approximated red and blue. • Reconstruct the red and blue channels using the complete green channel. 1. GRADIENTS 2. GROUPING 3. EDGE DETECT 3 pxls 4. EDGE DETECT 2 pxls 6. IMPROVED EDGE DETECTION 5. COLOR CORRELATION

  16. P7 P8 P9 P12 P13 P14 P17 P18 P19 Channel Reconstruction Background Green Channel Reconstruction Basics • For each Green pixel on a red center… Advanced Methods Channel Reconstruction Conclusion Bayer Pattern for Red Centered Pixel 1. GRADIENTS • A similar approach is taken to the finding the green value at a blue centered pixel 2. GROUPING 3. EDGE DETECT 3 pxls 4. EDGE DETECT 2 pxls 6. IMPROVED EDGE DETECTION 5. COLOR CORRELATION

  17. P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22 P23 P24 P25 Channel Reconstruction Background Blue and Red Channel Reconstruction Basics • Blue and red channels are then completed using the full green channel. Advanced Methods Channel Reconstruction Conclusion Where… Similar approach is taken for completing Red channel. Bayer Pattern for Red Centered Pixel 1. GRADIENTS 2. GROUPING 3. EDGE DETECT 3 pxls 4. EDGE DETECT 2 pxls 6. IMPROVED EDGE DETECTION 5. COLOR CORRELATION

  18. Conclusion Background Conclusion Basics Advanced Methods • Highly computational and hence slow. • Not suitable for real-time applications. Channel Reconstruction • Drastically reduces color artifacts. Conclusion • Improved Edge Quality. Thank You

  19. References Background Basics [1] D. Darian Muresan, S. Luke, and T. W. Parks, “Reconstruction of Color Images From CCD Arrays,” Cornell University, Ithaca NY. 1485, [2] R. Kimmel, “Demosaicing: Image Reconstruction from Color CCD Samples” IEEE Transl. J. Image Processing, vol. 8, Sept. 1999. [3] Xiaomeng Wang, Weisi Lin, Ping Xue, “Demosaicing with Improved Edge Direction Detection” IEEE Transl. J. Image Processing, 2005. Advanced Methods Channel Reconstruction Conclusion

More Related