Contrast enhancement is the baseline concept in the image processing. When digital image is processed in spatial domain, generally, the computations are performed on intensity values. Intensity value ranges from 0 to 255 for the 8-bit gray-level image. Being the numerical data is processed in gray-level images, the statistical inferences are, in general, drawn with respect to the desired problem to be solved. Presented work deals with the designing of hypothesis test for finding the given gray-level image is bright or dark. Once the inference is drawn, then the contrast enhancement is possible by using the known histogram equalization algorithms. In this paper, the hypothesis is formulated and tested for the image to find it as a candidate for the contrast enhancement or not. Analytical and experimental results are obtained to check the hypothesis testing, and it is found that the obtained results are encouraging.