A New Illumination Invariant Feature Based on SIFT Descriptor in Color Space

The SIFT descriptor is one of the most widely used descriptors and Is very stable in regard to change in rotation, scale, affine, illumination, etc. However, because of the greater emphasis on its insensitivity to geometric changes, this descriptor is weak in various illuminations. This method is based on key points extracted from the image. If there are many such points, a lot of time will be needed in the matching and recognition phases. Therefore, in this article, an attempt has been made in this article to use a normal color space both to use color information and to make the extracted features invariant against illumination variants. Moreover, a clustering technique has been used and similar key points have been eliminated in order to reduce the time needed to do the calculations. In other words, subtractive clustering has been used to choose those key points which are more distinct from and less similar to other points. In the section on results obtained, a successful implementation of this study is presented. The efficiency of the proposed algorithm and that of the base SIFT algorithm on the ALOI dataset were studied, and it was found out that by adding this method to the base SIFT descriptor, and with changes in illumination variants, the rate of recognition improves by 5% and that the calculation complexity also decreases.