Real-time imaging of CO in vehicle exhaust was demonstrated using a gas correlation spectrometry based mid-infrared camera for the first time. The novel gas-correlation imaging technique is used to eliminate the spectral interferences from background radiation and other major combustion products, and reduce the influences of the optical jitter and temperature variations, thereby identifying and quantifying the gas. We take several spectral factors into account for the instrument design, concentration calibration and data evaluation, including atmospheric transmission, radiation interference, as well as the spectral response of infrared camera, filter and gas cell. A calibration method based on the molecular spectroscopy and radiative transfer equation is developed to identify the numerical relationship between the CO concentration x length and the measured image intensity. Two-dimensional CO distribution of vehicle exhaust with a time resolution of 50 Hz and detection limit of 20 ppm x meter is achieved when the distance between optical equipment and engine nozzle is 3 m. The gas correlation spectrometry based mid-infrared camera shows a great potential as a future technique to monitor vehicle pollution emissions quantitatively and visually. (C) 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement.