Character extraction in digital images is an interesting topic in the field of image processing. One of the steps in this processes is image preprocessing, that is preparing the image before extraction. There are many methods that can be used, one of which is the WAN method. The WAN method is a development of the Sauvola method, a method commonly used to improve the quality of text images. In an journal article, the WAN method provides better results than other methods for improving the quality of text images. In this research, this claim was tested by applying it as a preprocessing process in text character extraction using Optical Character Recognition (OCR) technology. The purpose of this research is to find out how well the WAN method is used compared to other methods. The result is that OCR technology give better result to an original image compare to image with preprocessing process, including using WAN method with a match rate of 52.33%. Meanwhile, if the preprocessing process is carried out first, the WAN method gives better results than the Sauvola method with a match rate of 26.93% for the WAN method and 23.27 for the Sauvola method.