This paper proposes an efficient probabilistic indexing scheme called
(PR-Tree) to facilitate a efficient probabilistic retrieval of Chinese calligraphic manuscript images based on triple features such as contour points, character styles and number of strokes. To the best of our understanding, this is first work on probabilistic retrieval over Chinese character. Different from conventional character retrieval and indexing methods  which only adopts shape similarity as a query metric, our proposed indexing algorithm allows user to choose the above three kinds of features as query elements. Moreover, a probabilistic model is introduced into the character retrieval process. Comprehensive experiments are conducted to testify the effectiveness and efficiency of our proposed retrieval and indexing methods respectively.