Parents : Image Segmentation
| Date and time note was created | $= dv.current().file.ctime |
| Date and time note was modified | $= dv.current().file.mtime |
- Done using sharpening filters
- To get thicker lines we use First Order Derivative Filters.
- To get thinner lines we use Second Order Derivative filters .
The process steps are similar to Isolated Point Detection just the derivatives as seen above changes.
Steps in Isolated point detection
A threshold function would be given according to which we have to allocate pixel values in the end to form a binary image of the isolated detected points.
Link to original
- Apply zero padding to the image according to the kernel given.
- Now apply the kernel to the zero padded image and then replace the center with the calculated value.(Do this for entire image)
- Now the image obtained is compared with the threshold function and binary image is generated.
Filters/Kernels used according to direction of line.
Trick to remember the filters.
- Remember the horizontal filter
- Rotate by 90 degree to get verical filter
- In other filters only change the direction of position of the row with value 2 in direction specified.