While it was possible to come up with a sensor estimation from the data gathered, there were several innacuracies that contributed to error in the calculated functions. The first and most obvious is the experimental conditions. The light table, though useful for maintaining a standard illumination, did not provide a *uniform* illumination. In other words, there were brighter and darker patches, and this visually evident in the images from the digital camera. Secondly, there was an assumption made about color balancing, and the gamma correction functions that we removed from the data were only approximation. Even if the deviation from the true values had been small, the multiple levels of estimation would compound the errors.

One final hypothesis questions if there was one single color matching function that the camera used. We wondered if the designers had chosen arbitrary spans of the color spectrum (ie "reds") and chosen independent color matching functions that best approximated the data within that particular span. Regardless of reasons for errors though, it is obvious that the color matching functions derived through this experiment do provide a crude estimation of the actual CCD response of the Olympus digital camera.