Recent Posts

How to apply the Viterbi algorithm The goal of the Viterbi algorithm is find the most likely sequence of hidden states given some observed events. Lets say this is your HMM: A hidden Markov model (HMM) example We always start in $x$ and always end in $z$. Now you observed the sequence $O_1 = BAB$. What is … Read More »

How do I calculate a Histogram equalization? Let's say you have the following greyscale image: $A = \begin{pmatrix} 255 & 50 & 255\ 0 & 50 & 50 \end{pmatrix}$ Histogram Now the histogram is a function $H: [0,255] \rightarrow \mathbb{N}_0$. The histogram of $A$ is $H(x) := \begin{cases} 1 &\text{, if } x = 0\ 3 &\text{, if … Read More »

Zero Mean Normalized Cross-Correlation An image from Tsukuba University. This is one of hundreds of images that you can use to test your algorithms. Link is below. Zero Mean Normalized Cross-Correlation or shorter ZNCC is an integer you can get when you compare two grayscale images. Lets say you have a webcam at a … Read More »

k-nearest-neighbor classification and k-means - an interactive example When the circle has exactly the same number of blue / green dots in it, it will be green. When you move the mouse over the box, everything will be calculated and drawn again. This leads to flickering with k-means, as k-means includes a random choice of cluster centers. Changelog Version … Read More »

KogSys-Klausur Dieser Artikel beschäftigt sich mit der Vorlesung „Kognitive Systeme“ am KIT. Er dient als Prüfungsvorbereitung. Ich habe die Vorlesungen bei Herrn Dr. Waibel im Sommersemester 2013 gehört. Behandelter Stoff Vorlesung 15.04.2013 Kapitel 1 Einführung 17.04.2013   Faltung, Fouriertransformation, Dirac-Funktion 29.04.2013 Klassifikation I Schablonenanpassung: Probleme, Statistische … Read More »