1. Edges and Contours– Chapter 7

2. Visual perception We don’t need to see all the color detail to recognize the scene content of an image That is, some data provides critical information for recognition, other data provides information that just makes things look “good”

3. Visual perception Sometimes we see things that are not really there!!! Kanizsa Triangle (and variants)

4. Edges Edges (single points) and contours (chains of edges) play a dominant role in (various) biological vision systems –Edges are spatial positions in the image where the intensity changes along some orientation (direction) –The larger the change in intensity, the stronger the edge –Basis of edge detection is the first derivative of the image intensity “function”

5. First derivative – continuous f(x) Slope of the line at a point tangent to the function

6. First derivative – discrete f(u) Slope of the line joining two adjacent (to the selected point) point u-1u+1u

7. Discrete edge detection Formulated as two partial derivatives –Horizontal gradients yield vertical edges –Vertical gradients yield horizontal edges –Upon detection we can learn the magnitude (strength) and orientation of the edge More in a minute…

8. NOTE In the following images, only the positive magnitude edges are shown This is an artifact of ImageJ Process->Filters->Convolve… command Implemented as an edge operator, the code would have to compensate for this

9. Detecting edges – sharp image ImageVertical Edges Horizontal Edges

10. Detecting edges – blurry image ImageVertical Edges Horizontal Edges

11. The problem… Localized (small neighborhood) detectors are susceptible to noise

12. The solution Extend the neighborhood covered by the filter –Make the filter 2 dimensional Perform a smoothing step prior to the derivative –Since the operators are linear filters, we can combine the smoothing and derivative operations into a single convolution

13. Edge operator The following edge operators produce two results –A “magnitude” edge map (image) –An “orientation” edge map (image)

14. Prewitt operator 3x3 neighborhood Equivalent to averaging followed by derivative –Note that these are convolutions, not matrix multiplications

15. Prewitt – sharp image

16. Prewitt – blurry image

17. Prewitt – noisy image Clearly this is not a good solution…what went wrong? –The smoothing just smeared out the noise How could you fix it? –Perform non-linear noise removal first

18. Prewitt magnitude and direction

20. Sobel operator 3x3 neighborhood Equivalent to averaging followed by derivative –Note that these are convolutions, not matrix multiplications –Same as Prewitt but the center row/column is weighted heavier

21. Sobel – sharp image

22. Sobel – blurry image

23. Sobel – noisy image Clearly this is not a good solution…what went wrong? –The smoothing just smeared out the noise How could you fix it? –Perform non-linear noise removal first

24. Sobel magnitude and direction

26. Still not good…how could we fix this now? Using the information of the direction (lots of randomly oriented, non-homogeneous directions) can help to eliminate edged due to noise –This is a “higher level” (intelligent) function

27. Roberts operator Looks for diagonal gradients rather than horizontal/vertical Everything else is similar to Prewitt and Sobel operators

28. Roberts magnitude and direction

31. Compass operators An alternative to computing edge orientation as an estimate derived from two oriented filters (horizontal and vertical) Compass operators employ multiple oriented filters To most famous are –Kirsch –Nevatia-Babu

32. Kirsch Filter Eight 3x3 kernel –Theoretically must perform eight convolutions –Realistically, only compute four convolutions, the other four are merely sign changes The kernel that produces the maximum response is deemed the winner –Choose its magnitude –Choose its direction

33. Kirsch filter kernels Vertical edges L-R diagonal edges R-L diagonal edges Horizontal edges

34. Kirsch filter

35. Nevatia-Babu Filter Twelve 5x5 kernel –Theoretically must perform twelve convolutions –Increments of approximately 30° –Realistically, only compute six convolutions, the other six are merely sign changes The kernel that produces the maximum response is deemed the winner –Choose its magnitude –Choose its direction

36. Nevatia-Babu filter