Recognizing 3D objects using tactile sensing and curve invariants

Pattern Analysis and Applications, 2010

This paper presents a new method for recognizing 3D objects based on the comparison of invariants of their 2D projection curves. We show that Euclidean equivalent 3D surfaces imply affine equivalent 2D projection curves that are obtained from the projection of cross-section curves of the surfaces onto the coordinate planes. Planes used to extract cross-section curves are chosen to be orthogonal to the principal axes of the defining surfaces. Projection curves are represented using implicit polynomial equations. Affine algebraic and geometric invariants of projection curves are constructed and compared under a variety of distance measures. Results are verified by several experiments with objects from different classes and within the same class.

IEEE Transactions on Instrumentation and Measurement, 2002

A novel smart tactile sensor that recognizes the nature of the surfaces is presented. The approach is based on the idea of analyzing the signal produced when the sensor touches and stimulates the surface. An "intelligent probing" system for material recognition has been developed. It is based on the use of bimorph piezo-ceramic actuators and sensors that allow the unknown surface to be stimulated and the response signal sensed. Two different experimental prototypes of the tactile sensing system have been realized and their performance has been characterized. Several interesting applications have been considered with particular emphasis on the problems of "humanitarian demining" and automatic waste material recycling. Experimental results are given to show the efficiency of the smart measuring system. ).

2011

Abstract In recent years, autonomous robots have been increasingly deployed in unknown environments. In order to cope with the unknown, the capability to train autonomously the perception model of an environment is highly desirable. By developing proper sensing technology, this task can be significantly facilitated. In this paper, we explore the problem of artificial tactile perception, aimed at surface identification. To this end, we introduce a simple tactile probe based upon triple axis accelerometers.

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000

This paper discusses how data from multiple tactile sensors may be used to identify and locate one object, from among a set of known objects. We use only local information from sensors: 1) the position of contact points and 2) ranges of surface normals at the contact points. The recognition and localization process is structured as the development and pruning of a tree of consistent hypotheses about pairings between contact points and object surfaces. In this paper, we deal with polyhedral objects constrained to lie on a known plane, i.e., having three degrees of positioning freedom relative to the sensors. We illustrate the performance of the algorithm by simulation. Index Terms-Object recognition, robotics, tactile sensing.

Proceeding” CFT Varna, 2002

This paper is a part of the research on a project concerning the analysis of 3D contours. Namely, the problem of pattern recognition is becoming very actual, especially in connection to automation and robotization of industrial production. Therefore, this problem is considered in many mathematical disciplines: discrete mathematics, geometry, numerical mathematics, topology, combinatorics, computer science, etc. Besides various interesting mathematical contents, there are very important and rather concrete applications. All these led to the high activity in this field: publication of scientific articles and books, software development and construction of specialized facilities. A good review of this subject and the bibliography can be found for example in [5]. * This paper was presented at the workshop "Multiscale approximations (wavelets, splines, RBF) and applications to Image and Signal Processing" accompanying CTF 2002.

King Mongkut’s University of Technology North Bangkok International Journal of Applied Science and Technology, 2013

The authors have proposed a novel surface recognition algorithm capable of determining contact surfaces types by means of tactile sensor fusion since 2008. With Quadric surface representing; the limiting procedures which relate various degenerate Cartesian coordinate systems play a crucial result in the classification of all such systems. In this paper, the renew recognition processes are described. The space transformations are technically reducing the complicated contact surface for classification refinement. The complete classification of these modifiable coordinate systems is provided by means of the corresponding space curvature. Information is obtained directly at the interface between the object and the sensing device and relates to three-dimensional. The technique called “eigenvalue trajectory analysis”, is introduced after and adopted for specifying the margin of classification. The authors demonstrate mathematically approach which offers significant computational advantage...

Proceedings CVPR '89: IEEE Computer Society Conference on Computer Vision and Pattern Recognition

The space curves are highly descriptive features for 3-D objects. Invariant representations for space curves are discussed in this paper. We introduce a complex waveform representation for space curves. The waveform is parametrized by arc length. We also propose an invariant representation of space curves using the 3-D moment invariants of their breakpoints. Space curve matching using invariant global features is discussed. An algorithm for matching partially occluded 3-D curves is also presented, in which an associat,ion graph is constructed from local matchings. The maximal cliques of the graph will determine the longest portion of the model curves in the scene. This work was supported by the National Science Foundation under Grant IRI-8710856 and U.S. Army Research Office under Contract DAAL 0388K0033.

International Journal of Computer Vision, 2000

This article addresses the problem of recognizing a solid bounded by a smooth surface in a single image. The proposed approach is based on a new representation for two-and three-dimensional shapes, called their signature, that exploits the close relationship between the dual of a surface and the dual of its silhouette in weak-perspective images. Objects are modeled by rotating them in front of a camera without any knowledge of or constraints on their motion. The signatures of their silhouettes are concatenated into a single object signature. To recognize an object from novel viewpoint other than those used during modeling, the signature of the contours extracted from a test photograph is matched to the signatures of all modeled objects signatures. This approach has been implemented, and recognition examples are presented.

Computer Vision and Image Understanding, 1998

To recognize three-dimensional objects bounded by smooth curved surfaces from monocular image contours, viewpoint-dependent image features must be related to object geometry. Contour bitangents and inections along with associated parallel tangents points are the projection of surface points that lie on the occluding contour for a ve-parameter family of scaled orthographic projection viewpoints. An invariant representation can be computed from these image features and used for modeling and recognizing objects. Modeling is achieved by moving an object in front of a camera to obtain a curve of possible invariants. The relative camera-object motion is not required, and 3D models are not utilized. At recognition time, invariants computed from a single image are used to index the model database. Using the matched features, independent qualitative and quantitative veri cation procedures eliminate potential false matches. Examples from an implementation are presented.

2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2016

Tactile data and kinesthetic cues are two important sensing sources in robot object recognition and are complementary to each other. In this paper, we propose a novel algorithm named Iterative Closest Labeled Point (iCLAP) to recognize objects using both tactile and kinesthetic information. The iCLAP first assigns different local tactile features with distinct label numbers. The label numbers of the tactile features together with their associated 3D positions form a 4D point cloud of the object. In this manner, the two sensing modalities are merged to form a synthesized perception of the touched object. To recognize an object, the partial 4D point cloud obtained from a number of touches iteratively matches with all the reference cloud models to identify the best fit. An extensive evaluation study with 20 real objects shows that our proposed iCLAP approach outperforms those using either of the separate sensing modalities, with a substantial recognition rate improvement of up to 18%.