Surface deformation feature extraction based on 3D point cloud scan line object

The present invention discloses a kind of feature extraction based on the deformation object surface 3D point cloud scan line method. Including the use of the three-dimensional data acquisition based on the measuring sensor line structured light scanning, to obtain a three-dimensional point cloud data of the object surface data preprocessing; eliminate abnormal data, influence the texture extraction section for Main Profile, Main obtain accurate contour of the cross section of the object; binding knowledge deformation characteristics, based on the feature point extracting deformities section to obtain a binary image, an image based on the sub-block to obtain preliminary positioning deformed regions form sub-blocks of the target set; another modification feature points in the sub-block set for morphological operations, generating a confidence deformation region ROC, then using ROC geometric morphological characteristics, reduction target region growing, to ensure the integrity of the deformation detection region; final deformation characteristics according to a predefined statistical characteristic values ​​of the object surface deformation of the deformation region, so as to obtain accurate deformation of the object of the complete attribute information.