Search Results for: morphological operation

We run into problems when we take all of the pixels in a box around an input pixel and that pixel is close to one of the edges of the image. Let’s look at some strategies to deal with edge pixels.

Spatial operators are closely related to concepts from signal processing called correlation and convolution. They are similar but different and we discuss the important properties of convolution for Gaussian kernels.

Diadic operations involve two images of the same size and result in another image. For example adding, subtracting or masking images. As a realistic application we look at green screening to superimpose an object into an arbitrary image.

In mathematical morphology the values of the structuring element were either zero or one. The hit and miss transform extends this to also include a ‘don’t care’ value. We can use this transform to solve complex problems like finding the skeleton of an object or the intersection point of lines.

Let’s recall the key techniques we’ve covered including monadic and dyadic image processing operations and efficient ways to write these in MATLAB using vectorization.

For an image stored as a variable in the MATLAB workspace let’s look at how we access the values of individual pixels in an image using their row and column coordinates. Using the MATLAB colon operator we can extract an intensity profile, extract a submatrix which is a region of the image, flip the image […]

We learn how to describe the orientation of an object by a 2×2 rotation matrix which has some special properties. Try your hand at some online MATLAB problems. You’ll need to watch all the 2D “Spatial Maths” lessons to complete the problem set.

So far we have taken a linear combination of pixels in the box around the input pixel, but non-linear operations like sorting and ranking the pixel values also prove to be very useful. We look at the median filter which is much better at removing salt and pepper noise from image than simple smoothing.

Using the properties of convolution we can combine a simple derivative kernel with Gaussian smoothing to create a derivative of Gaussian (DoG) kernel which is very useful for edge detection, or a Laplacian of Gaussian (LoG) kernel which is useful for detecting regions.