1.3. Image processing tasks#
There are a plenty of possible ways to process an image. In the following we give a non-exhaustive list of examples for common image processing tasks.
Image denoising describes the removal of so-called Gaussian noise from images. It is probably the oldest and most studied problems in image processing.
Geometric denoising seeks to remove geometric noise (for example surface noise) while preserving the structure of the image.
Image deblurring describes the task of reducing the blur in images and making them appear sharper.
Image contrast enhancement is achieved by converting the color space of the image to enhance visible contrasts.
Image compression is a task aiming to reduce the space needed while maintaining most of the image’s structure/sharpness.
Image inpainting describes the objective of reconstructing missing regions in an image.
Colorization is the process of converting a grayscale (black and white) input image into a colorized output image that accurately represents the semantic colors and tones present in the input.
Image mosaicing is the process of blending together of several images to form one large image so that the boundaries between the original images are not visible.
Image warping describes the method of alternating already existing images to one’s liking.
Image segmentation lets us partition an image into different segments by distinctly differentiating between objects in the original image.
Motion detection is the process of detecting moving objects in temporal dependent images.