2 results about "Composite Position" patented technology

The invention discloses a point cloud completion method based on a space-time joint network, and the method comprises the steps: 1, fusing local angle coding, local coordinate coding and local feature information, and obtaining a multi-composite position coding feature; 2, constructing a double-branch adaptive Mama network comprising an adaptive local feature information module and a Mama global feature information module, realizing multi-level modeling of point cloud features, performing local-local and local-global combined learning, and fully mining the point cloud features; 3, generating a key point set by a seed point generator; 4, a double-branch adaptive Mama network and a neighborhood cross Transform are alternately used, and a complete point cloud is output; and 5, designing a loss function between the input incomplete point cloud and the output complete point cloud. According to the method, incomplete point cloud data caused by limitation of acquisition equipment or environment can be complemented, and the method adapts to different data sets and has good universality and robustness and excellent complementation performance.

This invention relates to a train hot axle monitoring method based on SRU-Hyena spatiotemporal dual-stream fusion, belonging to the field of train inspection technology. To address the technical challenges of limited accuracy and difficulty in uniformly modeling the spatiotemporal correlation of temperature sequences in traditional train hot axle monitoring methods under complex operating conditions, this invention obtains an enhanced axle temperature temporal tensor based on background temperature points, multiple axle temperature points, and statistical features of these points. An enhanced feature representation is obtained based on a composite position encoding layer and fed into two independent branches for processing. The temporal dynamic branch obtains temporal features through a multi-layer Simple Recurrent Unit (SRU), while the spatial correlation branch obtains spatial features through a multi-layer Hyena encoder based on long convolution and gating structures. A spatiotemporal fusion features are then obtained using a spatiotemporal feature interaction layer. Scaling and translation parameters are derived from a static conditional tensor, and affine transformations are performed on the spatiotemporal fusion features based on these parameters. Finally, a classifier is used to obtain the train hot axle monitoring results.