Fiber-reinforced polymer (FRP) composites are increasingly popular due to their superior strength to weight ratio. In contrast to significant recent advances in automating the FRP manufacturing process via 3D printing, quality inspection and defect detection remain largely manual and inefficient. In this paper, we propose a new approach to automatically detect, from microscope images, one of the major defects in 3D printed FRP parts: fiber-deficient areas (or equivalently, resin-rich areas). From cross-sectional microscope images, we detect the locations and sizes of fibers, construct their Voronoi diagram, and employ α -shape theory to determine fiber-deficient areas. Our Voronoi diagram and α -shape construction algorithms are specialized to exploit typical characteristics of 3D printed FRP parts, giving significant efficiency gains. Our algorithms robustly handle real-world inputs containing hundreds of thousands of fiber cross-sections, whether in general or non-general position.

The accuracy of fiber recognition in cross-sectional scanned images is critical to material characterization for Fiber-Reinforced Polymer (FRP) composites. It is challenging to accurately detect and locate all fibers, especially broken fibers. We propose a two-step approach to automatically detect and categorize fibers. First we exploit distance-transform-based watershed segmentation to extract the boundaries of individual fiber cross-sections to handle cases where fibers contact each other, then use our proposed contour gradient charts to evaluate the breakage of each fiber cross-section and classify individual fiber boundaries as circles or ellipses. Our method accurately and robustly recognizes fibers in the cross-sectional images, whether they are aligned or misaligned, complete or broken.

Selective Laser Melting (SLM) is a metal additive manufacturing technique. The lack of SLM process repeatability is a barrier for industrial progression. SLM product quality is hard to control, even when using fixed system settings. Thus SLM could benefit from a monitoring system that provides quality assessments in real-time. Since there is no publicly available SLM dataset, we ran experiments to collect over one thousand videos, measured the physical output via height map images, and applied a proposed image processing algorithm to them to produce a dataset for semi-supervised learning. Then we trained convolutional neural networks (CNNs) to recognize desired quality metrics from videos. Experimental results demonstrate our the effectiveness of our proposed monitoring approach and also show that the semi-supervised model can mitigate the time and expense of labeling an entire SLM dataset.

Most additive manufacturing processes fabricate objects by stacking planar layers of solidified material. As a result, produced parts exhibit a so-called staircase effect, which results from sampling slanted surfaces with parallel planes. Using thinner slices reduces this effect, but it always remains visible where layers almost align with the input surfaces. In this research we exploit the ability of some additive manufacturing processes to deposit material slightly out of plane to dramatically reduce these artifacts. We focus in particular on the widespread Fused Filament Fabrication (FFF) technology, since most printers in this category can deposit along slightly curved paths, under deposition slope and thickness constraints. Our algorithm curves the layers, making them either follow the natural slope of the input surface or on the contrary, make them intersect the surfaces at a steeper angle thereby improving the sampling quality. Rather than directly computing curved layers, our algorithm optimizes for a deformation of the model which is then sliced with a standard planar approach. We demonstrate that this approach enables us to encode all fabrication constraints, including the guarantee of generating collision-free toolpaths, in a convex optimization that can be solved using a QP solver. We produce a variety of models and compare print quality between curved deposition and planar slicing.

To address the problem of incorrect labels in training data for deep learning, we propose a novel and simple training strategy, Iterative Cross Learning (ICL), that significantly improves the classification accuracy of neural networks with training data that has noisy labels. We randomly partition the noisy training data into multiple separate subsets, each of which is used to train an independent network. After these independent networks predict labels for the original data, if the labels agree, we update the label with the predicted result for that data point, but otherwise we update the label with a random label, a key to the success of our proposed method. The process is repeated, possibly with several stages, to gradually improve the performance. Testing our method on MNIST and CIFAR-10 with partially shuffled labels, ICL significantly improves the classification accuracy of existing methods when the data labels have noise, especially in heavy noise situations. Moreover, the proposed method doesn't require any change to the underlying neural networks' structure or loss function, so it can also easily be combined with other existing methods that address noisy labels, improving their performance.

We present a novel descriptor algorithm (DUDE) using line/point duality and a randomization strategy that provides simple but robust, consistent feature extraction and correspondence. Using duality enables us to effectively capture a distribution of line segments, and the proposed randomization strategy improves repeatability over existing techniques by generating more line features in common between two images. We demonstrate the effectiveness of our approach using a challenging set of disparate image pairs, and show that the DUDE descriptor performs comparably to state-of-the-art methods with significantly less computation expense.

We propose a novel automated grading system that can compare two multiview engineering drawings consisting of three views that may have allowable translations, scales, and offsets, and can recognize frequent error types as well as individual drawing errors. We show that translation, scale, and offset-invariant comparison can be conducted by estimating the affine transformation for each individual view within drawings. Our system directly aims to evaluate students' skills creating multiview engineering drawings. Since it is important for our students to be familiar with widely used software such as AutoCAD, our system does not require a separate interface or environment, but directly grades the saved DWG/DXF files from AutoCAD. We show the efficacy of the proposed algorithm by comparing its results with human grading. Beyond the advantages of convenience and accuracy, based on our data set of students' answers, we can analyze the common errors of the class as a whole using our system.

We present practical algorithms for accelerating distance queries on models made of trimmed NURBS surfaces using programmable Graphics Processing Units (GPUs). We provide a generalized framework for using GPUs as coprocessors in accelerating CAD operations. By supplementing surface data with a surface bounding-box hierarchy on the GPU, we answer distance queries such as finding the closest point on a curved NURBS surface given any point in space and evaluating the clearance between two solid models constructed using multiple NURBS surfaces. We simultaneously output the parameter values corresponding to the solution of these queries along with the model space values. Though our algorithms make use of the programmable fragment processor, the accuracy is based on the model space precision, unlike earlier graphics algorithms that were based only on image space precision. In addition, we provide theoretical bounds for both the computed minimum distance values as well as the location of the closest point. Our algorithms are at least an order of magnitude faster and about two orders of magnitude more accurate than the commercial solid modeling kernel ACIS.

We present algorithms for computing accurate moments of solid models that are represented using multiple trimmed NURBS surfaces. Our algorithms make use of programmable Graphics Processing Units (GPUs) to accelerate the computations. We evaluate the surface coordinates and normals accurately, with theoretical bounds, using our GPU NURBS evaluator. We have developed a framework that makes use of this data to evaluate surface integrals of trimmed NURBS surfaces in real time. With our framework, we can compute volume and moments of solid models with theoretical guarantees. The framework also supports local geometry changes, which is useful for providing interactive feedback to the designer while the solid model is being designed. We can compute the center of mass and check for stability of the solid model interactively. Applications of such real-time moment computation include deformation modeling, animation, and physically based simulations.

We present a new approach for computing the voxelized Minkowski sum of two polyhedral objects using programmable Graphics Processing Units (GPUs). We first cull out surface primitives that will not contribute to the final boundary of the Minkowski sum. The remaining surface primitives are then rendered to depth textures along six orthogonal directions to generate an initial solid voxelization of the Minkowski sum. Finally we employ fast flood fill to find all the outside voxels. We generate both solid and surface voxelizations of Minkowski sums without holes and support high volumetric resolution of 10243 with low video memory cost. The whole algorithm runs on the GPU and is at least one order of magnitude faster than existing boundary representation (B-rep) based algorithms for computing Minkowski sums of objects with curved surfaces at similar accuracy. It avoids complex 3D Boolean operations and is easy to implement. The voxelized Minkowski sums can be used in a variety of applications including motion planning and penetration depth computation.

In injection molding, a flat parting surface that is normal to the mold parting direction achieves the best mold alignment with the least cost. However, for complex parts, parting surfaces that consist of a number of planes are necessary. In this paper, we provide an algorithm to find a near-optimal parting surface as a series of planes that intersect the boundary of the part.We form a continuous band of triangles on the part boundary that are parallel to the parting direction within a tolerance and perform vertical trapezoidation of the band. We can then find a set of planes that intersect vertical lines in the trapezoidation. We use a linear program to keep the planes normal to the parting direction.

We present practical algorithms for accelerating geometric queries on models made of NURBS surfaces using programmable Graphics Processing Units (GPUs). We provide a generalized framework for using GPUs as co-processors in accelerating CAD operations. By attaching the data corresponding to surface-normals to a surface bounding-box structure, we can calculate view-dependent geometric features such as silhouette curves in real time. We make use of additional surface data linked to surface bounding-box hierarchies on the GPU to answer queries such as finding the closest point on a curved NURBS surface given any point in space and evaluating the clearance between two solid models constructed using multiple NURBS surfaces. We simultaneously output the parameter values corresponding to the solution of these queries along with the model space values. Though our algorithms make use of the programmable fragment processor, the accuracy is based on the model space precision, unlike earlier graphics algorithms that were based only on image space precision. In addition, we provide theoretical bounds for both the computed minimum distance values as well as the location of the closest point. Our algorithms are at least an order of magnitude faster than the commercial solid modeling kernel ACIS.

Given a triangular mesh defining the geometry of a 3D workpiece filled with water, we propose an algorithm to test whether, for an arbitrary given axis, the workpiece will be completely drained under gravity when rotated around the axis. Observing that all water traps contain a concave vertex, we solve our problem by constructing and analyzing a directed “draining graph ” whose nodes correspond to concave vertices of the geometry and whose edges are set according to the transition of trapped water when we rotate the workpiece around the given axis. Our algorithm to check whether or not a given rotation axis drains the workpiece outputs a result in about a second for models with more than 100,000 triangles.

Existing haptic rendering algorithms are limited in the complexity of the models that can be simulated realistically without artifacts such as extraneous force discontinuities or pop-through. High haptic update rates for realistic feedback require fast collision detection methods, such as those achievable using distance fields. We introduce an efficient algorithm for reconstructing a C1distance field from discretely sampled distances and gradients. Our Coons-patch based interpolation method supports 50 kHz 3-DOF haptic rendering of complex models with high accuracy, while eliminating undesirable force discontinuities and pop-through. The approximation error of our representation on a 256×256×256 grid is similar to that of polygonal model reconstructions containing hundreds of thousands of polygons, an accuracy previously unobtainable with haptic force rendering.

We present algorithms for evaluating and performing modeling operations on NURBS surfaces using the programmable fragment processor on the Graphics Processing Unit (GPU). We extend our GPU-based NURBS evaluator that evaluates NURBS surfaces to compute exact normals for either standard or rational B-spline surfaces for use in rendering and geometric modeling. We build on these calculations in our new GPU algorithms to perform standard modeling operations such as inverse evaluations, ray intersections, and surface-surface intersections on the GPU. Our modeling algorithms run in real time, enabling the user to sketch on the actual surface to create new features. In addition, the designer can edit the surface by interactively trimming it without the need for retessellation. Our GPU-accelerated algorithm to perform surface-surface intersection operations with NURBS surfaces can output intersection curves in the model space as well as in the parametric spaces of both the intersecting surfaces at interactive rates. We also extend our surface-surface intersection algorithm to evaluate self-intersections in NURBS surfaces.

We present algorithms for evaluating and performing modeling operatyons on NURBS surfaces using the programmable fragment processor on the Graphics Processing Unit (GPU). We extend our GPU-based NURBS evaluator that evaluates NURBS surfaces to compute exact normals for either standard or rational B-spline surfaces for use in rendering and geometric modeling. We build on these calculations in our new GPU algorithms to perform standard modeling operations such as inverse evaluations, ray intersections, and surface-surface intersections on the GPU. Our modeling algorithms run in real time, enabling the user to sketch on the actual surface to create new features. In addition, the designer can edit the surface by interactively trimming it without the need for re-tessellation. We also present a GPU-accelerated algorithm to perform surface-surface intersection operations with NURBS surfaces that can output intersection curves in the model space as well as in the parametric spaces of both the intersecting surfaces at interactive rates.

This paper presents a new method to evaluate and display trimmed NURBS surfaces using the Graphics Processing Unit (GPU). Trimmed NURBS surfaces, the de facto standard in commercial 3D CAD modeling packages, are currently tessellated into triangles before being sent to the graphics card for display since there is no native hardware support for NURBS. Previous GPU-based NURBS display methods relied on first approximating the NURBS patches with lower degree Bezier patches before evaluation. Our method uses a GPU fragment program to evaluate the surface point coordinates of the original NURBS patch directly, from the control points and knot vectors stored as textures in graphics memory. This evaluated surface is trimmed during display using a dynamically generated trim-texture calculated via alpha blending. The implementation incorporates dynamic Level of Detail (LOD) for real-time interaction at different resolutions of the NURBS surfaces. We obtain rendering speeds at least one order of magnitude faster than evaluation using the CPU.

The contemporary engineering design curriculum embraces the concept of "concurrent engineering," in contrast to a more traditional, linear "over-the-wall" design process. In the old model, only after a design was all but complete would the manufacturing engineers become involved, typically requesting numerous design changes before compromising on a modified design that was both feasible and economical to manufacture. Not only does delayed manufacturability feedback increase time-to-market, it also increases cost. The flexibility to make design changes, particularly for interdependent components of a complex system, rapidly diminishes later in the design cycle. Since over 70% of manufacturing cost typically becomes fixed in the design stage, it is vital that designers be able to accurately assess the effect of design decisions on manufacturing costs early in the design process.

We present new programmable graphics hardware accelerated algorithms to test the castability of geometric parts and assist with part redesign. These algorithms efficiently identify and graphically display undercuts and minimum and insufficient draft angles. Their running times grow only linearly with respect to the number of facets in the solid model, making them efficient subroutines for our algorithms that test whether a tessellated CAD model can be manufactured in a two-part mold. We have developed and implemented two such algorithms to choose candidate directions to test for castability using accessibility analysis and Gauss maps. The efficiency of these algorithms lies in that they identify groups of candidate directions such that if any one direction in the group is not castable, none are, or if any one is castable, all are. We examine trade-offs between the algorithms' speed, accuracy, and whether they guarantee that a castable direction will be found for a part if one exists.