* If you want to update the article please login/register
For the first time, deep learning techniques can be used to reconstruct high-precision modulation distributions of specular surfaces from a single-frame fringe pattern under SMAT, allowing for quick and accurate defect detection of specular surfaces. This method can also be used to determine the 3D shape measurement in order to recover high-precision phase distributions of specular surfaces from a single-frame fringe pattern under PMD. We use depthwise separable convolution, residual structure, and U-Net to create an enhanced U-Net network in this paper.
Source link: https://doi.org/10.1364/oe.464452
Introduction: Shape segmentation is commonly used in several engineering disciplines to break a 3D shape into pieces for use in some specific applications. This paper reviews existing methods of shape segmentation in order to identify their pros and cons, as well as their pros and cons. Method: In two steps of feature extraction and model separation, the shape segmentation process is detailed. Conclusion: Clustering is the most commonly used method for shape segmentation. For identification, analysis, and reconstruction of large-scale models, machine learning techniques are leading the way.
Source link: https://doi.org/10.2174/1872212115666210203152106
* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions