Cris' Image Analysis Blog

Last month I posted a tutorial on snakes. I explained that with snakes, the segmentation solution is obtained by minimizing an energy functional. The gradient descent minimization then yields an update function for the snake that can be seen as a set of internal and external forces acting on the snake. I was curious to see where we would get if we were to define the snake simply as a set of balls, connected with each other through springs, moving over the image with momentum, friction, and all the rest. For this, we will completely abandon the notions of gradient descent, function minimization, and so forth, and write a simple Newtonian physics simulation.

(more…)

In this post I want to show how easy it is to implement snakes using MATLAB and DIPimage. The next release of DIPimage will have some functionality implementing snakes, and this post is based on code written for that. The first implementation of snakes I made for DIPimage used a custom class, dip_snake, to simplify usage. We decided eventually to do without the class. The code discussed here uses that first implementation, which you can download here. The most important bits of the code are identical to that in the next release of DIPimage, but the wrapping is not. Adding the class is a nice excuse to show how to modify DIPimage to understand a new parameter type. I will discuss that in a future post.

(more…)

Edit May 2018: Since publishing this post, the MATLAB Image Processing Toolbox has added the function imgaussfilt that correctly applies a Gaussian smoothing filter. For Gaussian derivatives, the recommendations here still apply.

If you don’t use DIPimage, you probably use MATLAB’s Image Processing Toolbox. This toolbox makes it really easy to do convolutions with a Gaussian in the wrong way. On three accounts. The function fspecial is used to create a convolution kernel for a Gaussian filter. This kernel is 2D. That’s the first problem. The other two problems are given by the default values of its parameters. The default value for the kernel size is [3 3]. The default value for the σ (sigma) is 0.5. (more…)

In my recent lectures on filtering I was trying to convey only one thing to my students: do not use the uniform filter, use the Gaussian! The uniform (or “box”) filter is very easy to implement, and hence used often as a smoothing filter. But the uniform filter is a very poor choice for a smoothing filter, it simply does not suppress high frequencies strongly enough. And on top of that, it inverts some of the frequency bands that it is supposed to be suppressing (its Fourier transform has negative values). There really is no excuse ever to use a uniform filter, considering there is a very fine alternative that is very well behaved, perfectly isotropic, and separable: the Guassian. Sure, it’s not a perfect low-pass filter either, but it is as close as a spatial filter can get.

Because recently I found some (professionally written) code using Gaussian filtering in a rather awkward way, I realized even some seasoned image analysis professionals are not familiar and comfortable with Gaussian filtering. Hence this short tutorial.

(more…)