Machine vision users are beginning to consider color imaging an important source of information for quality assurance and part verification. Applications in lumber sorting, food inspection, digital film conversion and print inspection are among those benefiting from color cameras and related technology. Conventional silicon photodetectors cannot distinguish colors, so filters separate an image into its red, green and blue components. High-performance industrial camera designers have developed three ways to get color from monochromatic linear image sensors: two-chip, three-chip and trilinear techniques. The two-chip method is not common because it requires color interpolation software, which some say does not create true colors. The three-chip method is a common technique for both area- and line-scan cameras. The trilinear method is a traditional line-scan camera technique. All three styles provide the user with RGB data. Which is the best technique for your application? Photonics Spectra asked two camera manufacturers -- Basler Vision Technologies of Exton, Pa., and Dalsa Inc. of Waterloo, Ontario, Canada -- to explain the application trade-offs between the most common line-scan color-separation techniques.