Image contrast enhancement plays an important role in many machine vision applications. In this study, we propose a new method for edge enhancement based on analysis on embedded surfaces of images. The proposed method gives an insight into the relationship between the image intensities (also the gradients) and image contrast. In our method, scaled surface area and the surface volume are used to reconstruct the image for edge enhancement, and then the contrast of the reconstructed image is adjusted by a 'strengthen-weaken' process. Although, current method for edge enhancement such as curvelet transform can enhance the edge part, it does not provide good tonal rendition or color constancy sometimes. The experimental results show that our method can give good performance not only in edge enhancement, but also in tonal rendition and color constancy.