In many fields such as telemedicine, electronic commerce, image archive of ruins and Internet mail order, color reproduction technique plays a key role in capturing an accurate image of target. The color of a target can be faithfully reproduced by presuming and recording the spectrum reflectance of the object which is not influenced by the lighting conditions using a multiplespectrum camera. However, this technique has a disadvantage that a lot of money is required. Since it is cheaply realizable, a method of reducing the influence of the lighting conditions using a known color chart is used widely. Unfortunately, it is necessary to input the color chart position in the picture into a color reproduction system manually. In order to reduce this troublesome work, we proposed a method of estimating a color chart position in a picture. Traditionally, there is a technology that detects objects with SIFT (Scale-Invariant Feature Transform) that obtained features from local images in object detection. SIFT has robust features which are invariant to image scaling, and rotation, illumination changes. Using SIFT we can compute the similarity transformation between a template color chart picture and a target picture. We show the effectiveness of our method by experiments using real images under various lighting conditions.