Recognition of carrier flight deck operations based on multi-dimensional features

To address the challenges brought by unique flight operational scenarios and insufficient public data for carrier flight deck operations, this study proposes a recognition method based on multi-dimensional features.

First, key points such as deck passage boundaries and static obstacles are accurately selected to represent the environmental information. Interactions between dynamic operational participants and static deck facilities are modelled using graph convolutional networks to explore their underlying connections of deck operation interaction relationships. Then, a multi-scale spatio-temporal feature extraction (MS-STFE) module is designed, incorporating a dilated attention mechanism that captures key individual interactions at both global and local levels by applying different dilation rates. At the same time, temporal convolutional networks (TCN) combined with the attention mechanism are employed to extract temporal interaction features, efficiently capturing dynamic relationships across both long and short sequences. Finally, the MS-STFE module is stacked multiple times to adaptively extract multi-dimensional features, thereby improving the recognition accuracy of carrier flight deck operations.

Experiments conducted on a self-constructed dataset featuring multi-perspective carrier flight deck operation scenarios involving heterogeneous deck operation entities demonstrate that the proposed method significantly outperforms existing group activity recognition models such as ARG, DIN, AT, and GroupFormer, achieving an accuracy of 97.8%.

This study provides a valuable reference for the high-accuracy recognition of carrier flight deck operations.