Texture is an essential characteristic of the visual patterns appearing in natural surfaces and images. The local binary pattern has a wide variety of applications ranging from texture classification (histopathological images, remote sensing images to image segmentation (biomedical imaging). Many proposed feature descriptor methods that are easy to implement have low performance while other methods are not able to improve even by modifying using various transforms like translation, rotation, affine, and perspective transform. Local binary patterns (LBP) have come forth as one of the extensively studied descriptors for texture type. Although extensive research has been carried out using LBP in industrial inspection, facial recognition, character recognition, it, however, remains open for further work mainly in the medical area. Paper first describes texture classes in detail with a special focus on LBP. A comparative analysis of various rotations is performed on a medical data set. provides a detailed description of Local binary patterns and comparative analysis of various rotation invariant variants of LBP in terms of classification accuracy of medical images.