Old pictures restoration and enhancement

Old pictures records scenes and features of historical events. Due to aging, improper handling, and changes of chemicals used in producing the image, these images deteriorate. This paper reports the results of an ongoing study of applying image processing methods in the restoration and enhancement of old pictures. The process is not straightforward because of many difficulties both in the image itself and in the method and the format used to transfer the image into the computer. The process of image recording produces a noise and the format used to record the image also affects the final result. The type and model of noise on these old pictures is unknown. This paper studies the noise characteristics. The most important result obtained proposes a signal dependent noise model for the noise in old pictures. This result is important in identifying the proper method to be used in noise filtering. The paper also applies and compares the results of some traditional and modified image filtering methods. Traditional filtering methods such as low pass averaging, median filtering and adaptive filtering are used. Results are presented to illustrate the effect of these filtering methods.