Oxygen saturation is an indicator of oxygen transport in the body, and indicates if sufficient oxygen is being supplied to the body, especially to the skin and surface. Based on a new proposed algorithm of NIR image correction of the arterial oxygen saturation (SpO2), we have developed a two-dimensional oxyhemoglobin saturation measurement which is demonstrated by using the combination technique of multispectral imaging and a proposed calibration method. Dual-spectral images are acquired under the uniform illumination at the wavelengths for 633 and 940 nm to derive the images diffusely reflected light from the skin tissue. For calibrating the superficially reflected light from the skin, the skin tissue and a proposed diffusive material for reflection comparison are all illuminated by LED array light. After the 2D reflection images are taken, a new algorithm of oxygen saturation images are developed to calibrate and build two dimension information of oxygen saturation. The haemoglobin oxygen saturation levels corresponding to all image pixels are estimated by the parameters derived from the proposed calibration method with respect to calibration conditions of different oxygenated haemoglobin. We successfully develop a new calibration algorithm by using the charge-coupled device (CCD) camera to receive the images. The system adopts two proper red and near-infrared wavelengths to derive the iris tissue blood oxygen saturation and investigate their correlation for medical diagnosis.