To address the issue of point cloud noise affecting the accuracy of external parameter calibration for different LiDARs, an adaptive joint calibration method for LiDAR and camera based on DBSCAN is proposed. In this method, firstly, pass-through filtering is used to pre-process point clouds to obtain checkerboard corner points, secondly, KDTree is constructed to obtain the density threshold within the radius of corner points. Finally, based on the obtained density threshold, DBSCAN algorithm is used to cluster the complete checkerboard point clouds, thus avoiding the problem of time consuming and precision reduction caused by the inconsistency of different LiDAR external parameter calibration methods. The experimental results show that the efficiency and accuracy of the method are improved, the calibration efficiency is increased by 4-6 times, and the reprojection error is reduced by about 29.8%. The resulting external parameters are suitable for engineering applications.