Geometric Feature Matching Research Articles

Laser in-situ measurement in robotic machining of large-area complex parts

High-precision manufacturing of key components in aerospace plays a crucial role in determining their operational performance, and shape measurement during machining is essential for controlling component accuracy. This article proposes a laser in-situ high-precision measurement method based on improved ICP registration during robotic machining to resolve the contradiction between measurement accuracy and range. Firstly, a robotic in-situ measurement system based on line laser scanning is constructed. Subsequently, the FM-BiKD-ICP algorithm based on geometric feature matching and bi-directional KD-tree proposed in this article is used to complete the high-precision registration of multi-million-scale point cloud data of complex parts with large surface area. Finally, the standard ball target ruler is measured, with the RMSE is under 0.079 mm after multiple registrations, and the average error of hole center distance is less than 0.037 mm on large-area complex parts. These results indicate the high precision of the proposed robot in-situ measurement system.

Intelligent Vehicle Inspection Tool Design Based on Freeman Chain Code for Automatic Annotation of 3D Models

Autonomous vehicle are more and more widely used in daily life, and the requirements for their safety performance are higher and higher. As a tool for testing auto parts, intelligent inspection tools are crucial to the guarantee of automobile quality. However, traditional fixture design relies on manual drawing, which is inefficient and prone to errors. To solve this problem, this research uses Freeman chain code to determine the annotation object, uses case clustering method to annotate, and uses error back propagation algorithm to realize case knowledge classification learning, and designs intelligent vehicle inspection tool design technology based on Freeman chain code 3D automatic annotation method. The experimental results show that the geometric feature matching results are correct, and the difference in feature comparison results is significant, with a high accuracy rate. Meanwhile, the geometric similarity annotation method has a high accuracy rate, taking only 3 minutes to complete the annotation, which is 7 minutes longer than traditional manual annotation. The error backpropagation algorithm can accurately achieve feature classification, and the design time of size chain inspection tool deformation design is reduced by 214 minutes compared to manual reverse deformation design, significantly improving design efficiency. In summary, the proposed design method for automotive inspection tools can achieve automatic model annotation, improve design efficiency, and reduce design time.

Precise plane registration with multiple geometric feature matching and weighted least squares estimation

Point cloud registration is a fundamental problem for 3D laser scanning technology, which is extensively applied in geographic entity modelling such as 3D reconstruction of urban roads and buildings. Registration accuracy is one of the main focuses for these applications. However, noisy points, limited overlap, varying data sources, and differing measuring accuracy may cause changes between point cloud sets and reduce registration accuracy. This paper introduces an automatic plane registration method for urban roads and building scenes, which does not need manual on-site deployment. Firstly, plane primitives are extracted using voxel-based filtering region growth. Next, corresponding planes for the extracted primitives are identified by leveraging saliency features and constructing adjacency matrices. Finally, plane registration is achieved using a weighted plane coordinate conversion model. Through real-world scene experimentation, an overall accuracy of 10 cm and a segmental registration accuracy of 5–6 cm is achieved with our method, outperforming both feature point-based and global point cloud registration approaches in terms of efficiency and accuracy.

The Study and Application of Facial Recognition Models

Facial recognition technology is a biometric application that uniquely identifies or authenticates a person by comparing and analyzing facial features based on the width of a person's face. Face recognition technology has gained popularity recently thanks to the development and invention of artificial intelligence and other technologies. The traditional methods of face recognition technology development—the method based on geometric features and the method based on 3D models—as well as the deep learning Convolutional Neural Network (CNN) model VGG-Face for face recognition—are analyzed, compared, and described in this paper. It also provides a detailed description of face recognition technology. In this work, accuracy and TAR values are employed for geometric feature matching by comparing the metric results of each method on a dataset for better comparison. Our results show that while both traditional geometric feature modeling methods and 3D modeling methods can achieve good accuracy, however, they do not perform as well as humans in terms of accuracy. The Vgg-Face based on the deep learning model completely exceeds the 3D model method in terms of accuracy and TAR value, and is even more accurate than humans.

LIO-CSI: LiDAR inertial odometry with loop closure combined with semantic information.

Accurate and reliable state estimation and mapping are the foundation of most autonomous driving systems. In recent years, researchers have focused on pose estimation through geometric feature matching. However, most of the works in the literature assume a static scenario. Moreover, a registration based on a geometric feature is vulnerable to the interference of a dynamic object, resulting in a decline of accuracy. With the development of a deep semantic segmentation network, we can conveniently obtain the semantic information from the point cloud in addition to geometric information. Semantic features can be used as an accessory to geometric features that can improve the performance of odometry and loop closure detection. In a more realistic environment, semantic information can filter out dynamic objects in the data, such as pedestrians and vehicles, which lead to information redundancy in generated map and map-based localization failure. In this paper, we propose a method called LiDAR inertial odometry (LIO) with loop closure combined with semantic information (LIO-CSI), which integrates semantic information to facilitate the front-end process as well as loop closure detection. First, we made a local optimization on the semantic labels provided by the Sparse Point-Voxel Neural Architecture Search (SPVNAS) network. The optimized semantic information is combined into the front-end process of tightly-coupled light detection and ranging (LiDAR) inertial odometry via smoothing and mapping (LIO-SAM), which allows us to filter dynamic objects and improve the accuracy of the point cloud registration. Then, we proposed a semantic assisted scan-context method to improve the accuracy and robustness of loop closure detection. The experiments were conducted on an extensively used dataset KITTI and a self-collected dataset on the Jilin University (JLU) campus. The experimental results demonstrate that our method is better than the purely geometric method, especially in dynamic scenarios, and it has a good generalization ability.

3DTDesc: learning local features using 2D and 3D cues

Pairwise frame registration with sparse geometric local features on real-world depth images is not particularly robust due to the low resolution and incomplete nature of the 3D scan data. Moreover, there might be many regions with similar geometric information. In this paper, we present 3DTDesc, a data-driven descriptor which closely combines both 2D texture and 3D geometric information for frame registration. The proposed descriptor is learned directly from color point clouds, which is time-efficient and provides robust and accurate geometric feature matching in a variety of settings. The texture information and the geometric information closely interact in the fusing network, which are complements of each other in situations of textureless regions or regions with similar geometric information and different texture information. We also propose a multi-scale 3DTDesc to further improve the performance of the feature matching. The effectiveness and efficiency of our proposed 3DTDesc are demonstrated by extensive experimental results on challenging RGB-D datasets and various ablation studies.

Learning to fuse local geometric features for 3D rigid data matching

This paper presents a simple yet very effective data-driven approach to fuse both low-level and high-level local geometric features for 3D rigid data matching. It is a common practice to generate distinctive geometric descriptors by fusing low-level features from various viewpoints or subspaces, or enhance geometric feature matching by leveraging multiple high-level features. In prior works, they are typically performed via linear operations such as concatenation and min pooling. We show that more compact and distinctive representations can be achieved by optimizing a neural network (NN) model under the triplet framework that non-linearly fuses local geometric features in Euclidean spaces. The NN model is trained by an improved triplet loss function that fully leverages all pairwise relationships within the triplet. Moreover, the fused descriptor by our approach is also competitive to deep learned descriptors from raw data while being more lightweight and rotational invariant. Experimental results on four standard datasets with various data modalities and application contexts confirm the advantages of our approach in terms of both feature matching and geometric registration.

Ranking 3D feature correspondences via consistency voting

This paper addresses the 3D correspondence ordering problem by proposing a consistency voting (CV) method. Given a set of plausible feature correspondences generated via geometric feature matching, our goal is to rank these correspondences accurately and robustly such that reasonable correspondences can be located according to their rankings. The proposed method employs a voting-based scheme, which is based on the consistency check of each initial correspondence with a predefined voting set, to calculate a voting score for each correspondence. We consider two consistency terms, i.e., rigidity and local reference frame (LRF) affinity, to ensure a correct judge when two correspondences are compatible. LRF is an intrinsic canonical cue widely used in many existing local geometric features. The repeatability of LRFs helps to remove the ambiguity that sometimes arises from the rigidity check. The proposed CV method is evaluated on three public benchmarks including both LiDAR and Kinect data. Comparison with the state-of-the-art confirms the effectiveness and robustness of the CV method with respect to noise, varying point densities, clutter, occlusion, and changes in data modality. Furthermore, the proposed CV method can rank thousands of correspondences in real time.

ICE VELOCITY MEASUREMENT IN EAST ANTARCTICA FROM 1960s TO 1980s BASED ON ARGON AND LANDSAT IMAGERY

Abstract. Ice flow velocity is a vital parameter for estimating the ice mass balance of glaciers in Antarctica. Especially long time serial observation of the surface velocity is of great significance to assessing the relationship between Antarctic ice materials and global climate change. However, the existing research on Antarctic ice velocity based on remote sensing data since 1970s due to the harsh climate in Antarctica. This paper presents an ice flow velocity estimating method includes image pre-processing, geometric model reconstruction, image ortho-rectification and feature matching by using ARGON images token in 1963 and Landsat images collected form 1973 to 1989.Considering the temporal-spatial distributions of ARGON images and Landsat images in Antarctica, two different methods respectively based on ortho-photos pair and Non-Ortho photos are adopted in this paper. More specifically, when there exist two stereo pairs taken in different time in the glacier region, after being ortho-rectified, the stereo pairs can be used to calculate ice flow velocity based on feature matching method. Otherwise, a parallax decomposition method that separates the effect of the terrain relief from the ice flow motion is applied when there only exists one stereo pair with a certain time interval. With this method, glacier surface velocity is available in the glacier region lacked enough stereo pairs. The methods mentioned above for estimating ice flow velocity are applied in Totten, Amery and Fimbul, etc. in eastern Antarctica. Furthermore, a 1960-80s ice flow speed map in the main glaciers of East Antarctica is produced for the first time.

VECTOR CLUSTERING OF PASSIVE MILLIMETER WAVE IMAGES WITH LINEAR POLARIZATION FOR CONCEALED OBJECT DETECTION

Passive millimeter (MMW) imaging can penetrate cloth- ing to create interpretable imagery of concealed objects. However, the image quality is often restricted by low signal to noise ratio and temperature contrast as well as low spatial resolution. In this paper, we explore a four-channel passive MMW imaging system operating in the 8 and 3mm wavelength regimes with linear vertical and horizon- tal polarization directions. Both registration between difierent channel images and segmentation of concealed objects are addressed. Multi- channel image registration is performed by geometric feature matching and a-ne transform, and then multi-level segmentation separates the human body region from the background, and concealed objects from the body region, sequentially. In the experiments, several metallic and non-metallic objects concealed under clothing are captured in indoors. It will be shown that our method can separate objects with higher accuracy than the conventional method.

3D Model Based Pose Invariant Face Recognition from a Single Frontal View

This paper proposes a 3D model based pose invariant face recognition method that can recognize a face of a large rotation angle from its single nearly frontal view. The proposed method achieves the goal by using an analytic-to-holistic approach and a novel algorithm for estimation of ear points. Firstly, the proposed method achieves facial feature detection, in which an edge map based algorithm is developed to detect the ear points. Based on the detected facial feature points 3D face models are computed and used to achieve pose estimation. Then we reconstruct the facial feature points’ locations and synthesize facial feature templates in frontal view using computed face models and estimated poses. Finally, the proposed method achieves face recognition by corresponding template matching and corresponding geometric feature matching. Experimental results show that the proposed face recognition method is robust for pose variations including both seesaw rotations and sidespin rotations.

Direct cortical mapping via solving partial differential equations on implicit surfaces

In this paper, we propose a novel approach for cortical mapping that computes a direct map between two cortical surfaces while satisfying constraints on sulcal landmark curves. By computing the map directly, we can avoid conventional intermediate parameterizations and help simplify the cortical mapping process. The direct map in our method is formulated as the minimizer of a flexible variational energy under landmark constraints. The energy can include both a harmonic term to ensure smoothness of the map and general data terms for the matching of geometric features. Starting from a properly designed initial map, we compute the map iteratively by solving a partial differential equation (PDE) defined on the source cortical surface. For numerical implementation, a set of adaptive numerical schemes are developed to extend the technique of solving PDEs on implicit surfaces such that landmark constraints are enforced. In our experiments, we show the flexibility of the direct mapping approach by computing smooth maps following landmark constraints from two different energies. We also quantitatively compare the metric preserving property of the direct mapping method with a parametric mapping method on a group of 30 subjects. Finally, we demonstrate the direct mapping method in the brain mapping applications of atlas construction and variability analysis.

Optimum Matching of Geometric Features for Material Metamorphosis in Heterogeneous Object Modeling

AbstractThis paper presents a new methodology of material blending for heterogeneous object modeling by matching the material governing features for designing a heterogeneous object. The proposed geometric feature matching method is used to generate non-self-intersecting and non-twisted connecting surfaces. The proposed method establishes point-to-point correspondence represented by a set of connecting lines between two material directrices. To blend the material features between the directrices, a heuristic optimization method developed with the objective is to maximize the sum of the inner products of the unit normals at the end points of the connecting lines and minimize the sum of the lengths of connecting lines. By subdividing the connecting lines into equal number of segments, one can construct a series of intermediate piecewise curves that represent the material metamorphosis between the governing material features. Implementation and examples are also presented.

Augmented geophysical data interpretation through automated velocity picking in semblance velocity images

Velocity Picking is the problem of picking velocity-time pairs based on a coherence metric between multiple seismic signals. Coherence as a function of velocity and time can be expressed as a 2-D color semblance velocity image. Currently, humans pick velocities by looking at the semblance velocity image; this process can take days or even weeks to complete for a seismic survey. The problem can be posed as a geometric feature matching problem. A feature extraction algorithm can recognize islands (peaks) of maximal semblance in the semblance velocity image: a heuristic combinatorial matching process can then be used to find a subset of peaks which maximizes the coherence metric. The peaks define a polyline through the image, and coherence is measured in terms of the summed velocity under the polyline and the smoothness of the polyline. Our best algorithm includes a constraint favoring solution near the median solution for the local area under consideration. Each image is first processed independently. Then, a second pass of optimization includes proximity to the median as an additional optimization criterion. Our results are similar to those produced by human experts.

A General Method for Geometric Feature Matching and Model Extraction

Popular algorithms for feature matching and model extraction fall into two broad categories: generate-and-test and Hough transform variations. However, both methods suffer from problems in practica...

Robust affine structure matching for 3D object recognition

We consider model-based object localization based on local geometric feature matching between the model and the image data. The method is based on geometric constraint analysis, working in transformation space. We present a formal method which guarantees finding all feasible matchings in polynomial time. From there we develop more computationally feasible algorithms based on conservative approximations of the formal method. Additionally, our formalism relates object localization, affine model indexing, and structure from multiple views to one another.