Copy move forgery is a popular image tampering technique that copies part of the image onto another region of the same image. Detection is possible by exploring similar regions in an image and based on regional features' similarity. Many feature extraction algorithms are used to extract information from non overlapping blocks of the image. Similar features from separate regions are an indication of a copy move forged image. Image moments are used to represent the image blocks during forgery detection recently. Krawtchouk moments are used to extract regional features and detect copy move forgery in this paper. They are not used to extract features and detect forged images before. Experimental results indicate that the proposed method can detect copy move forgery for both regular and irregular shaped regions. Besides, the method is resilient to additive white Gaussian noise, blurring attacks. Experimental results also show that the method has higher accuracy ratio compared to similar works for post processed (Gaussian blurred) forged images.