DESCRIPTIONPerforms edge detection using two Sobel kernels. The module performs two independent filtering operations using convolution kernels denoted X and Y, then adds the absolute values of the outputs according to
The output is clipped to the valid data range for its word type, for instance [0..255] for byte data and [0..65535] for short data. The two kernels areoutput = | convolve( input, X ) | + | convolve( input, Y ) |
Edge detection by convolution with Sobel kernels is discussed in:X Y -1 +0 +1 +1 +2 +1 -2 +0 +2 +0 +0 +0 -1 +0 +1 -1 -2 -1
Digital Image Processing, Gonzales, R.C., Wintz, P., Addison Wesley, Second Edition, 1987, p 337.
Port: Img In
Port: Img Out
Edge enhanced filtered image.
© The Numerical Algorithms Group Ltd, Oxford UK. 2000