This paper proposes a novel technique, based on Canonical correlation analysis, and the Chi-square test is employed to test the significance of the correlation coefficients. If it is significant, then it is concluded that the input query and target images are same or similar; otherwise, it is inferred that the two images are significantly different. In order to experiment the proposed canonical correlation method, a database is designed and constructed with the help of different types of images and their feature vectors. The Fβ-measure is applied to evaluate the performance of the proposed method. The obtained results expose that the proposed technique yields better results than the existing.