You can use this tool to control your cookie settings. Otherwise, we will assume you are OK to continue.
MIL includes tools to help identify parts, products, and items using color, assessing quality from color as well as isolating features using color. The Color Distance tool reveals the extent of color differences within and between images. The Color Projection tool separates features from an image based on their colors and can also be used to enhance color to grayscale conversion for subsequent analysis using other grayscale tools. The Color Matching tool determines the best matching color from a collection of samples for each region of interest within an image.
A color sample can be specified either interactively from an image—with the ability to mask out undesired colors—or using numerical values. A color sample can be a single color or a distribution of colors (i.e., histogram). The color-matching method and the interpretation of color differences can be manually adjusted to suit particular application requirements. The Color Matching tool can also match each image pixel to color samples to segment the image into appropriate elements for further analysis using other tools.
MIL includes color-relative calibration to correct color appearance due to differences in lighting and image sensing, thus enabling consistent performance over time and across systems. Three methods are provided: Histogram-based, sample-to-sample, and global mean variance. The first method is unsupervised, only requiring that the reference and training images have similar contents. The second method is semi-supervised, requiring the correspondence between color samples on reference and training images, typically of a color chart. The third method is best suited for dealing with color drift and relies on global color distribution.