The measurement of laser scanner in different bands is affected by the reflection characteristics of atmospheric conditions, target surface and other factors, which makes the data cross-sensitivity between bands, resulting in data exhibiting uncertainty in the time series, thus diminishing the classification accuracy of the laser data. For this reason, a multi-band laser data classification considering time series fuzzy segmentation is proposed. Firstly, combined with principal component analysis and genetic algorithm, the features of multi-band laser data are extracted, the maximum feature vector is found out, and the time series of multi-band laser data is constructed. Secondly, the genetic algorithm is employed to optimize the laser data sequence, which mitigates the impact of of cross sensitivity on the classification results. Then, the fuzzy segmentation algorithm is utilized to divide the optimized laser data sequence into several time series segments, and in conjunction with the K-means algorithm, sequence segment clustering is completed to achieve precise classification of multi-band laser data and enhance its classification accuracy. The experimental results show that this method yields high classification accuracy and excellent classification effects when applied to multi-band laser data classification.