Design of Dynamic Grouping Algorithm for Sports Track Competition Based on Deep Learning

Due to the change of type, accuracy and spatial position of multi-sensor measurement system in sports track, the optimal measurement accuracy cannot be obtained during numbers fusion. A new dynamic grouping fusion algorithm is proposed. The traditional way of processing is inefficient. In the commonly used sports meeting administration software, sports track events are divided into eight fixed lanes, and the scheduling algorithm is fixed, which has certain defects. The application of cloud computing technology can effectively solve the efficiency problem in the processing of dynamic grouping algorithms, and it adopts parallel computing to accelerate computing tasks. The method used in this paper is deep learning. Deep learning is two types of methods that are distinguished by referring to the learning method of the learner and the comprehension ability they want to achieve. Both of them have significant differences in learning attitude and knowledge composition. Deep learning is a kind of high-end learning that achieves the realm of learning through the use of deep learning methods to achieve higher learning results, and such learning results can improve personal thinking ability.