Color images are stored as size 3 by width by height where the 3 refers to the RGB (red., green, blue) dimensions. For a lot computer vision problems color isn't important and the size of the image can be reduced to 1 dimension that is grayscale. This reduces the space and computational complexity of the model in training.

University of Michigan - Ann Arbor

Acquired data is usually messy and can come from different sources, angles, color scales, etc. In order to use machine and deep learning techniques they need to be standardized. Usually the goal is to increase the accuracy while reducing the complexity of the model used. There is no single answer to preprocessing but there are a group of techniques commonly chosen from.

Computer Vision Pre-Processing

Converting to grayscale

For certain machine learning algorithms it is necessary to resize a dataset of images to a unified dimension. This means scaling the images so they have identical widths and heights. This can be done using interpolation or down sampling.

Standardize images

Involves augmenting the dataset with scaled, rotated, decolorized, etc versions of the original image. This makes it so your model is invariant to a wide variety of variations of an image. Makes it more likely that it'll recognize objects when they appear in any form.

Learn Before

Related