Image segmentation to be basic for image analysis and recognition process. Segmentation divides the image into several regions based on the unique homogeneous image pixel. Image segmentation classify homogeneous pixels basedon several features such as color, texture and others. Color contains a lot of information and human vision can see thousands of color combinations and intensity compared with grayscale or with black and white (binary). The method is easy to implement to segementation is clustering method such as the Fuzzy C-Means (FCM) algorithm. Features to beextracted image is color and texture, to use the color vector L* a* b* color space and to texture using Gabor filters. However, Gabor filters have poor performance when the image is segmented many micro texture, thus affecting the accuracy of image segmentation. As support in improving the accuracy of the extracted micro texture used method of Local Binary Patterns (LBP). Experimental use of color features compared with grayscales increased 16.54% accuracy rate for texture Gabor filters and 14.57% for filter LBP. While the LBP texture features can help improve the accuracy of image segmentation, although small at 2% on a grayscales and 0.05% on the color space L* a* b*.