The hand-held 3 D laser scanner is affected by a variety of error factors, which makes the collected point cloud and the accuracy of the final measurement results, so the result needs to evaluate the uncertainty of the measurement results. According to the working principle of the scanner, the source analyzed the uncertainty, establish the uncertainty model and evaluate GUM, MCM and adaptive AMCM. The uncertainty evaluation results of the measurement plate planness experiment show that the MCM method fully considers the distribution type of input quantity, and the uncertainty evaluation result is more accurate than the GUM method; the AMCM method can effectively balance the contradiction of sampling number M too large or too small, and the process of measurement uncertainty evaluation is more convenient and efficient than the GUM and MCM methods.