The invention relates to an intelligent human femur X-ray film reading method, which comprises the following steps of carrying out denoising pretreatment on a skeleton X-ray film to enable an image tobe smooth, and enhancing an image contrast ratio through image sharpening; rapidly extracting the contour characteristics of the key part of the femoral joint in the X-ray film through a self-adaptive radiation gradient method; utilizing a scale invariant feature transformation algorithm to perform feature extraction on the bone X-ray film image; carrying out characteristic matching through an improved BP neural network, outputting a diagnosis result, and completing the intelligent reading of the human body femur X-ray film. The method has the advantages of being high in robustness and adaptive iteration, the human skeleton X-ray film is intelligently read through the scale invariant feature transformation algorithm and the improved BP neural network technology, the accuracy and the highefficiency are achieved, and the dependence of X-ray film diagnosis on manual reading of doctors is reduced to the maximum degree.