Finger Knuckle recognition system and finger back surface is one of the recent biometric methods used to identify people. The finger knuckle print is highly rich in textures and can be used to uniquely identify a person. In this paper we propose figure knuckle based biometric person identification using subspace methods such as Gabor filter, PCA and LDA. These methods need two images: test image and train image. Firstly, crop the test image and train image. The cropped images are changed to black and white image, then preprocessing with gabor filter to enhance the FKP image. After using gabor filter, feature extraction is employed by using PCA. The features of test image and train image are compared by using LDA. Then the output will be displayed
