A new algorithm based on Shape Context(SC) and Principal Component Analysis(PCA)called PCA-SC was proposed to improve the matching efficiency and anti-noise performance in shape matching and object recognition. The algorithm establishes a covariance matrix based on the feature matrix obtained by the SC, then reduces its dimensions according to the size of eigen value and forms a new feature matrix to implement the shape matching and object recognition. The proposed algorithm can not only remove noise interference and improve the recognition accuracy, but also can enhance the matching efficiency for real-time application. The experimental results of MNIST database indicate that the PCA-SC algorithm outperforms previous SC algorithm, and its recognition speed is doubled that of SC and the accuracy reaches to 96.15% increased by 0.5%. Furthermore, the anti-noise performance becomes stronger. Therefore, this algorithm shows better performance for shape matching and object recognition in efficiency, accuracy and anti-noise.