Aiming at the situation that the motion recognition of sports athletes is interfered by a variety of factors and the recognition results are not ideal, this paper uses the maximum spanning tree algorithm as the model basis to use machine learning ideas to construct a sports player motion recognition model based on the maximum spanning tree algorithm. Moreover, this article combines a region growing algorithm based on simple surface fitting and morphological reconstruction to initially segment sports actions. After that, this paper improves the prim algorithm, and combines an optimized watershed segmentation framework to construct a new energy function using the T-prim minimum spanning tree algorithm proposed in this paper. The constructed T-prim tree is combined with this optimized watershed segmentation framework to complete the segmentation of sports images.Finally, this paper designs experiments to verify the actual effect of the method proposed in this paper. It can be seen from the research results that the model constructed in this paper basically achieves the expected goal.

针对体育运动员的动作识别受多种因素干扰，识别结果不理想的情况，本文以最大生成树算法为模型基础，运用机器学习思想构建体育运动员的动作识别。基于最大生成树算法的模型。此外，本文结合了基于简单表面拟合和形态重构的区域增长算法，以初步分割运动动作。之后，本文对prim算法进行了改进，并结合优化的分水岭分割框架，利用本文提出的T-prim最小生成树算法构造了新的能量函数。构造的T-prim树与该优化的分水岭分割框架相结合，完成了运动图像的分割。最后，本文设计了实验，验证了本文提出的方法的实际效果。从研究结果可以看出，本文构建的模型基本达到了预期目的。