An efficient method on noise type recognition is proposed based on gray relational information of histogram, which can distinguish three familiar types of image noise, i.e., salt-and-pepper noise, Gaussian noise and speckle noise. In the method, statistical information of polluted image histogram is analyzed first and salt-and-pepper noise is then distinguished out directly. Gray relational analysis theory is employed to analysis the histogram curve characteristic of Gaussian noise and speckle noise, the two types of noise are recognized according to the different gray relational grades. Experimental results indicate that this method may provide with correct information on noise types when there is a subimage where the gray scales of all pixels are the same or very similar.

image noise; histogram; noise type; recognition; gray relational analysis (GRA)