Filters
Any operation that transforms an image into another image
What is a filter?
Predicting which category a new data point belongs to
What is classification?
A mathematical structure which relates input data and output data and is created from a number of data points in a set
What is a model?
Method of noise reduction that works by calculating the average of pixel values around a pixel
What is blur/moving average/gaussian filter?
Algorithm that connects neighbors and assigns labels to those regions of the image
What is bwlabel?
Whether two pixels are neighbours or not
What is connectedness?
Classifying a new data point based on the closest training data points
What is nearest neighbour classification?
Fitting a line to data points using the error in y-axis only
What is least squares?
Reconstructing 3D data from a series of images
What is computer vision?
Method for combining multiple close data points or regions
What is clustering?
A graph showing how frequently every pixel value appears in an image
What is a histogram?
Predicting one out of two classes
What is Binary Classification?
Fitting a line to data points using the error in both the x-axis and the y-axis
What is total least squares?
Removing all pixels above a certain value
What is thresholding?
Attaching category labels to regions within an image
What is semantic segmentation?
A simple type of transformation that transforms pixels from one value to another
What is gray level transform?
A theorem that describes the relation between the posterior probability, the prior probability and the likelihood
What is Bayes theorem?
Robust fitting method that reduces the influence of outliers by fitting to many small subsets of the data and choosing the fit that works best for the whole dataset
What is RANSAC?
Method for reducing noise in an image without blurring the edges
What is anisotropic filtering/structure adaptive smoothing?
Algorithm that aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest mean
What is k-means?
The properties of a data point or image, often used in classification
What are features?
A classification method that operates by transforming the data points and dividing them using hyperplanes
What is support vector machine?
Model fitting method that takes into account all the models that every feature or data point is compatible with
What is voting?
Image processing pperation that is sensitive to specific shapes in the input image. Each pixel in the image is adjusted based on the value of other pixels in its neighborhood of this shape. Examples are "open" and "close".
What are Morphological Operations?
Image segmentation method that divides the image by turning the image into a graph and solving a maximum flow problem
What is max-flow min-cut?