![]() In this paper we compare thirteen different methods for converting from color-to-grayscale. Indeed, color may be of limited benefit in many applications and introducing unnecessary information could increase the amount of training data required to achieve good performance. The main reason why grayscale representations are often used for extracting descriptors instead of operating on color images directly is that grayscale simplifies the algorithm and reduces computational requirements. A priori, none of these criteria suggest superior recognition performance. The most common techniques are based on weighted means of the red, green, and blue image channels (e.g., Intensity and Luminance), but some methods adopt alternative strategies to generate a more perceptually accurate representation (e.g., Luma and Lightness ) or to preserve subjectively appealing color contrast information in grayscale images (e.g., Decolorize ). ![]() However, since many methods for converting to grayscale have been employed in computer vision, we believe it is prudent to assess whether this assumption is warranted. This is because most researchers assume that the color-to-grayscale method is of little consequence when using robust descriptors. Modern descriptor-based image recognition systems often operate on grayscale images, with little being said of the mechanism used to convert from color-to-grayscale.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |