When using femtosecond laser to perform micro-scale ablation on monocrystalline silicon materials,the derived plasma luminescence spot contains a lot of processing information.How to effectively extract the geometric feature information of the spot image is of great significance to the study of the femtosecond laser ablation process.The profile feature is one of the most important features to describe the laser spot.This paper focuses on the analysis of the relationship between the profile feature of the plasma spot and the law of the spot movement and the classification method of the spot movement.Firstly,considering the weak light spot,the light spot image is enhanced.Secondly,the chain code method is adopted to extract the contour features of the light spot,and a small number of descriptors are selected to reconstruct the contour features of the light spot image by utilizing the low-frequency features of the Fourier descriptors,which reduces the complex computation and effectively improves the speed and quality of the light spot image classification.Finally,Hu invariant moment feature is used to prove the effectiveness of Fourier descriptors in extracting the contour features of laser spot.