Surface remeshing is widely used in computer vision and reverse engineering. The vertex normal vectors used in the existing surface remeshing methods are not accurate enough in the characteristic regions such as large curvature change, boundary and sharp edge, which leads to poor remeshing results in these regions. In this paper, the vertex normal vector based on hybrid weight and boundary correction is introduced to our method. We use this normal vector to improve the reprojection step in vertex translation, and combining edge splitting, collapsing, and flipping, propose an improved real-time adaptive remeshing method. Three models are selected for remeshing experiments. Compared with the RAR (real-time adaptive remeshing) method based on Thürrner's vertex normal, the Hausdorff distance is reduced by 4.99%, 8.79% and 6.85% respectively in the uniform length remeshing, and 3.21%, 2.25% and 11.92% respectively in the adaptive length remeshing. The results show that the remeshing effect obtained by our method is smoother and has higher geometric similarity with the original model.
Mesh Boolean is a basic operation in mesh modeling and the remeshing method is the key technology. This paper proposed an adaptive remeshing method with the hybrid of local edge length and curvature information, for high quality and efficient mesh Boolean in multi-resolutions. Firstly, calculate the local control length according to local edge length and curvature information. Then split too long edges, collapse too short edges, flip edges to minimize vertex valance, relax vertex position in tangent plane, project it back to the initial mesh. Iterate upper five steps to complete remeshing. Finally, to prove the efficiency and better control of mesh resolutions of the proposed algorithm, compare it with two art-to-date remeshing methods, using the models of dataset Thingi10k and apply it to an adaptive mesh Boolean method. The result shows that the average length is 55% shorter in high curvature area and 78% longer in low curvature area compared with uniform remeshing.
Defect detection for specular surfaces plays a vital role in precision manufacturing. However, traditional defect detection methods are unsuitable for specular surfaces because of their specular reflection property. The defect detection on specular surfaces is usually performed by inspectors, which makes the defect detection a time-consuming and unstable task. Deflectometry has been widely used in defect detection for specular surfaces combined with machine learning. Nevertheless, conventional deflectometry methods use the local curvature deviation map based on the unwrapped phase, which can only detect geometrical defects. Moreover, hand-crafted features need to be defined for each specific task. We present a method based on deflectometry and deep learning. Deflectometry provides the input images for the network, and the deep learning network completes the identification and location of defects. In deflectometry, the proposed method uses the light intensity contrast map to replace the local curvature map, which can detect both geometrical and textural defects. Based on conventional networks, depthwise separable convolution kernel is applied to reduce parameters, and residual convolution block is utilized to alleviate vanishing or exploding gradients. A subnet for feature aggregation is used to obtain multiscale information of defect features. Performance evaluation based on experiment results proved the effectiveness of the proposed method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.