Feature extraction process may include multiple steps and employ variety of image processing methods and techniques with the purpose to extract features of interest from the image bitmap.
Segmentation is the processing of the digital raster image with intention to separate the areas of the image that carry information of interest from the rest. The result of such operation usually is a binary mask that separates the areas of interest from the rest of the image. Number of numerical processing techniques may be used to achieve segmentation such as aperture filters, treshholding, edge detection, Fourier Transform and other frequency-based filtering etc.
Feature Detection and Object Recognition are tasks of discrimination and identification of the object of interest by specific set of attributes. Prime examples include machine vision, face recognition, license plate identification and other applications
The result of feature extraction is usually a binary mask that identifies the features of interest on the image