Nanjing University of Aeronautics and Astronautics

Anyi Huang

Jun Wang

Qian Xie

Zhongde Shan

Yuanpeng Liu

Zikuan Li

Cheng Yi

Zeyong Wei

Hao Yan

Zhoutao Wang

The present disclosure provides a point cloud denoising method based on multi-level attention perception, including the following steps: constructing a data set of point cloud denoising; constructing a point cloud denoising neural network, including a patch feature encoder, a global level perception module, a global level attention module, and a multi-offset decoder module, and training a network model by using the data set of point cloud denoising; for input point cloud, separately obtaining a neighborhood patch of a point of each original data point, and inputting coordinates of each data point in the neighborhood patch of a point to a trained denoising neural network to obtain a location offset of each original point; and separately adjusting, based on the obtained location offset, a location corresponding to each original data point in the input point cloud, to complete point cloud denoising.

Point cloud denoising method based on multi-level attention perception

A method for automatically processing a structure-reinforcing member of an aircraft, including: (S) acquiring, by a handheld laser scanner, data of an area to be reinforced of the aircraft; (S) controlling a robotic arm to automatically grab the reinforcing member for automatic scanning; (S) setting a cutting path in a computer aided design (CAD) digital model followed by registration with real data to obtain an actual cutting path, and cutting the reinforcing member; (S) controlling the robotic arm to guide a cut reinforcing member to a scanning area for automatic scanning; and (S) subjecting point cloud data of the cut reinforcing member and the area to be reinforced to virtual assembly and calculating a machining allowance to determine whether an accuracy requirement is met; if yes, ending a task; otherwise, grinding the reinforcing member automatically, and repeating steps (S)-(S).

Method for automatically processing structure-reinforcing member of aircraft

A method for detecting and recognizing surface defects of an automated fiber placement composite based on an image converted from point clouds, including: acquiring a surface point cloud of the automated fiber placement composite; fitting a plane to surface point data; calculating a distance from each point of the surface point cloud to a fitted plane; enveloping the surface point cloud by OBB, and generating a grayscale image according to the OBB and the distance; constructing a pre-trained semantic segmentation network for defect of fiber placement, and inputting the grayscale image to segment and recognize defect areas thereon; mapping a segmentation result output by the semantic segmentation network to the point cloud followed by defect evaluation and visualization.

Method for detecting and recognizing surface defects of automated fiber placement composite based on image converted from point cloud

A method for visualizing a large-scale point cloud based on normal, including: (S) according to a spatial structure of a point cloud data, constructing a balanced octree structure of a node point cloud; (S) according to the balanced octree structure and normal information of a point cloud, constructing an octree structure with the normal information; and constructing a normal level-of-detail (LOD) visualization node through downsampling; and (S) determining a node scheduling strategy according to a relationship between a viewpoint, a viewing frustum and a normal of a render node; and respectively calling a reading thread and a rendering thread to simultaneously perform reading and rendering according to the node scheduling strategy.

Growing community of inventors

Nanjing, China

Anyi Huang