Abstract:Point cloud classification for substation is challenging due to the occluding layout as well as the complexity of methods. Most previous methods trade inference speed for precision by building complex extractors. To alleviate the paradox of performance and complexity trade-off, this paper proposes the lightweight long-range attention classification for substation point cloud, including shuffle channel attention(SCA) and dilated channel attention(DCA). First, SCA captures local cross-channel interaction via 1D convolution and global interaction via channel shuffling, shown promising performance. Second, to further reduce the amount of computation involved in shuffling, we propose a more elegant method DCA by resizing the 1D vector after pooling into 2D feature map. Note that proposed DCA are implemented by just 2D convolution, determining the coverage of cross-channel interaction, with a significantly higher inference speed. Furthermore, we develop a heuristic algorithm to adaptively determine parameters like kernel size, shuffle group and size of 2D feature map. Besides ModelNet40 and ScanObjectNN(PB_T50_RS), this paper also selects ten of main objects in substation for training and testing. Finally, experimental results suggest that proposed methods bring notable performance gain, nearly 1%. DCA reaches 93.801% and just increases 0.0003M parameters but 15.5% faster than SCA. We also make an attempt to simplify proposed methods via reducing the coding dimensions and coding blocks. And simplified DCA achieves 93.4% performance with almost quarter parameters and 112% faster, guaranteeing both efficiency and effectiveness. A simplified version is used to achieve the highest accuracy of 85.115% and 83.304% on ScanObjectNN, with the highest improvements of 1.735% and 2.391%. This paper also conducts robustness test where different proportions of points are missing. Results show that when 93.75% missing, the accuracy only decreases by 4.093%, and can still reach 89%, which is significantly ahead of other methods.

Study of Base Segmenting Algorithm of Substation Equipment Based on 3D Point Cloud

3D point cloud recognition of substation equipment based on plane detection

Power Equipment Segmentation of 3D Point Clouds Based on Geodesic Distance with K-means Clustering

Research on Point Cloud Power Line Segmentation and Fitting Algorithm

A New Approach of Point Cloud Processing and Scene Segmentation for Guiding the Visually Impaired

An Improved Supervoxel Clustering Algorithm of 3D Point Clouds for the Localization of Industrial Robots

An Integrated Approach for Accurate 3D Point Cloud Recognition of Substation Equipment Using PSO and kNN Classification

Recognition of Three-dimensional Substation Equipment Based on Hough Transform

3D Surface Segmentation from Point Clouds Via Quadric Fits Based on DBSCAN Clustering

Semantic Segmentation of Substation Site Cloud Based on Seg-PointNet

An Improved Defect Detection Method for Substation Equipment

Surface Extraction and Boundary Detection Based on DBSCAN Clustering in 3D Point Clouds

Adaptive Clustering for Point Cloud

3D Object Segmentation Using Cross-Window Point Transformer with Latent Semantic Boundary Guidance

GEOMETRIC ATTRIBUTE ANALYSIS AND SEGMENTATION OF POINT CLOUD

A model-driven approach for fast modeling of three-dimensional laser point cloud in large substation

An 3D Point Cloud Reconstruction Approach from Single Image for Smart Power Plant

A Point Cloud Segmentation Algorithm Using Local Convexity and Octree

An Object Segmentation Method for Railway Surrounding Environment Based on Three-dimensional Point Cloud

Point cloud segmentation based on Euclidean clustering and multi-plane extraction in rugged field

Long-range attention classification for substation point cloud