Abstract: Stellar catalogue matching is a fundamental task in astronomical observations based on CCD images. In this paper, a stellar catalogue matching algorithm based on k-d tree and k-means clustering algorithm is proposed in Python. Triangle invariant tuples are used for blind matching of similar triangles based on Astroalign, which can indirectly calculate the scale of CCD image. In this paper, k-d tree is used for three times to optimize the calculation, and k-means clustering algorithm is used to segment the image to improve the matching precision. For testing, we match the stars with both sparse field frames taken by the 1 m telescope at Yunnan Observatory and dense field frames taken by the 2.4 m telescope at Yunnan Observatory. Our results show that the method can effectively adapt to subtle change of image scale and improve matching precision.