In this paper, we propose a novel and effective method for improving the accuracy of separating reflection components in a single image based on the dichromatic reflection model after calculating the diffuse reflection component by any existing method. Separating reflection components accurately is very important and useful in computer vision to detect highlight areas, which are often regarded as outliers, and to enhance image quality, especially texture, because we can control the intensity and apply a filter independently to each reflection component. Our proposed method is based on unsharp masking and an energy function. The method does not require any information in advance, and is applicable to many products, such as TVs. Experimental results show that our method can improve the separation of reflection components by a maximum of 13.06 dB in the peak signal-to-noise ratio (PSNR).