Denoise filters: Median

Gauss and Mean filters required a good knowledge of mathematics to understand its background. Median filter is very similar to both of them, but it based in a more logical approach than a mathematical one.

Denoise filters: Mean

Continuing with article series about the denoise filters, we had seen that the Gauss filter had a very high sensitivity to noise due to it's static nature. Now is time to analyze a dynamic denoise filter based in the arithmetic mean of the input pixels, this is called the Mean Denoise Filter. This filter adapts it's convolution kernel according to the values of the input pixels.

Denoise filters: Gauss

We have seen before about convolution filters, and at the end of that article we have seen a kernel used for removing noise in an image, that is random spurious pixels in a image, the process of removing that random pixels is called Denoise. In this case I will show you the general form of the denoising kernel based in gaussian distribution.

Integral image and image blur

In this post I will show you an image processing technique that provides a great boost to algorithms that requires the summation of a large number of pixels, this technique is called integral image, also known as Summed Area Table (SAT). Additionally, I will show you an algorithm used to blur an image that take advantage of the integral image.