A new classification method for terrestrial LiDAR point cloud intensity is proposed, which uses the color information of point cloud to constrain the intensity classification. Compared with the existing method solely based on intensity classification, the proposed method uses the color-to-strength supplement to establish the fault-tolerant mechanism of the laser intensity correction results, thus improving the problem of poor classification resulted from the fact that the intensity correction model cannot get the best correction result. The data of intensity and color information of the Faro Focus 3D 120 terrestrial laser scanner are investigated in the experiment. The results indicate that the proposed method can improve the accuracy of the three-dimensional point cloud data classification and can also improve the credibility and reliability of classification results even when the intensity data are poorly calibrated.