object-orientedprogramming using C++ classes is established practice in the general programming community and is beginning in computer music applications (Chaudhary, Freed et al. 1988;Chaudhary 1998). However, large ...
详细信息
the paper considers a statistical method to classify objects based on sound emitted by the objects. Engines and the carriageable devices are the sources of signal. the used methodology doesn't require the detailed...
详细信息
ISBN:
(纸本)0780345037
the paper considers a statistical method to classify objects based on sound emitted by the objects. Engines and the carriageable devices are the sources of signal. the used methodology doesn't require the detailed analysis which part of object is responsible for the components of signal. the sound is preprocessed using DSP methods to obtain feature vectors to be used by a statistical classifier. A computer system equipped by DSP processor for mobile object recognition as real-time application is also presented. Results of classification of various military vehicles were the subject of conclusions.
Using a computer vision system to supervise a construction task requires the reliable recognition of assembled objects. A particularly reliable object recognition is possible if a comprehensive representation of knowl...
详细信息
ISBN:
(纸本)0780345037
Using a computer vision system to supervise a construction task requires the reliable recognition of assembled objects. A particularly reliable object recognition is possible if a comprehensive representation of knowledge about assembled objects is provided. In this paper we present a cyclic semantic network which models all possible assemblies of objects from a wooden construction-kit for children. Due to its simplicity this model allows a straight recognition of assembled objects.
the Common object Request Broker Architecture is a successful, standardized system integration framework based on distributed object technologies. An ongoing effort concerns extensions to CORBA to incorporate real-tim...
详细信息
ISBN:
(纸本)0818686464
the Common object Request Broker Architecture is a successful, standardized system integration framework based on distributed object technologies. An ongoing effort concerns extensions to CORBA to incorporate real-time computing needs. Although the use of objects in veal-time computing is straightforward, the technical challenge lies in the replacement of static real-time computing infrastructures with a flexible real-time computing infrastructure, in which distributed real-time client and sewer objects can be created and connected as needed during runtime. We propose the concept of "Composite objects" for the integration of real-time and non-real-time computing into a single object-based framework. Within the paper, we present a methodology for creating objects which interface to CORBA without violating real-time assumptions. We present a example scenario which integrates a real-time computing architecture and CORBA components via "Composite objects". Finally, we discuss implementation-related issues based on our experiences with "Composite objects".
this paper concentrates on object models for the recognition of complex three-dimensional objects with a robot vision system. After giving a short overview on existing approaches, some demands on object models for rob...
详细信息
ISBN:
(纸本)0780345037
this paper concentrates on object models for the recognition of complex three-dimensional objects with a robot vision system. After giving a short overview on existing approaches, some demands on object models for robot vision systems are formulated. Afterwards, an approach of hybrid object models that fulfills all of these demands is presented. these hybrid models integrate neurobiologically motivated object representations by modell neurons similar to complex cortical cells and the explicit representation of objects by semantic networks, a well known methodology in the field of symbolic artificial intelligence. thereby, we can combine the attribute of robustness and fault tolerance of neural networks withthe well structured design of symbolic processing. Additionally, the procedural interface of our semantic networks allows the development of active vision systems and the implementation of reliable recognition on the basis of multiple viewpoints.
In this paper a study is presented how self-organizing ART networks can be used to create a trainable, feature based real time 3d object recognition system. Feature extraction is a well known approach to reduce the nu...
详细信息
ISBN:
(纸本)0780345037
In this paper a study is presented how self-organizing ART networks can be used to create a trainable, feature based real time 3d object recognition system. Feature extraction is a well known approach to reduce the number of appearances of a three-dimensional object. Since features are derived from only a small part of the information comprised in the original image it can not be assumed that a given set of objects is separable in the reduced feature space. To avoid ambiguities, in general multiple features have to be integrated in an object recognition system Since feature extraction can be computationally intensive a real time system should evaluate features sequentially and terminate recognition when ambiguities are resolved. this paper gives an analysis of the clustering properties of ART 2A-E networks. It is shown how ART networks can be used to generate meaningful hints concerning the object's identity from ambiguous features by exploiting them up to an optimal degree.
Scene analysis in robotics is divided into camera calibration, image scanning, determination of relevant information (a graph representation), object recognition and determination of positions for grasp planning. Ofte...
详细信息
ISBN:
(纸本)0780345037
Scene analysis in robotics is divided into camera calibration, image scanning, determination of relevant information (a graph representation), object recognition and determination of positions for grasp planning. Often grasp planning is working well, when the conditions are optimal. But in many cases there are errors caused by bad illumination, e.g. this results in an incorrect image description. this article presents some new ideas to solve this problem. Specially we concentrate on the problem of 3D-object recognition. For roughly choosing those models and pose parameters in a time-optimal way we use model-based invariants which make it possible to use a priori information. As a quality measure for the so-called "fine selection" we combine paint to set distances and gradient information. In what follows we give a description of the theoretical background and we discuss problems in practical applications. We finish with experiments and applications.
In the field of railway industry, more and more producers use extruded aluminium profiles as basic elements for the manufacturing of wagon crates. A mayor advantage compared to traditional manufacturing methods is the...
详细信息
ISBN:
(纸本)0780345037
In the field of railway industry, more and more producers use extruded aluminium profiles as basic elements for the manufacturing of wagon crates. A mayor advantage compared to traditional manufacturing methods is the good potential for automated workcells and thus reduced production costs, even considering the small batches found in this area. Due to the deviations between the optimal CAD-workpiece on which the programming of the robot is based and the real profile, it is necessary to correct the path of the robot. the following paper presents a new solution of integrating a sensor into an existing production process for milling window openings. For the milling of these openings the optimal movement of the robot describes a plane rectangle. Aspects which are discussed in this paper include the general approach which has been followed to develop the sensor system, the mathematical procedures for calculating the optimal path and finally the interfacing between the sensor system and the robot.
this paper presents a vision system (object localization and recognition) for a grasping robot environment. Our approach to object localization is based on the sequential integration of early vision processes, such as...
详细信息
ISBN:
(纸本)0780345037
this paper presents a vision system (object localization and recognition) for a grasping robot environment. Our approach to object localization is based on the sequential integration of early vision processes, such as color and edge detection. No assumptions about the object or background are necessary for this process. It detects blobs of interest in the scene and treats them as object candidates. the method presented here shows great reliability, flexibility and robustness. When the localization process completes, the next task is to classify the object in terms of its shape, orientation, size and exact position, as a basis for grasping. Recognition is achieved by comparing the image to stored two-dimensional object views. Stored views are represented as labeled graphs and are derived automatically from images of object models. Graph nodes are labeled by edge information, graph links by distance vectors in the image plane. Graphs emphasize occluding boundaries and inner object edges. these are identified by extracting local maxims in the Mallet wavelet transform of the image. Stored graphs are compared to test images by elastic matching. the system is robust with respect to surface markings and cluttered background. Our experiments demonstrate that the system is capable of fairly reliable object recognition and pose estimation in natural scenes.
programming robotic applications is hardly standardized. A major reason for this is the very special hardware equipment applied to the robot and the aim the software is written for. As a result, a modular approach inc...
详细信息
ISBN:
(纸本)0780345037
programming robotic applications is hardly standardized. A major reason for this is the very special hardware equipment applied to the robot and the aim the software is written for. As a result, a modular approach including easy maintenance and reusability of the modules is difficult to realize withthe systems currently available on the market. OSCAR therefore has been designed as a general purpose concept to create modular robotic software written in C++. It declares four software layers, each boundary representing a well defined step of abstraction from hardware related to high level related data and services. Data and services are handled by OSCAR's task management without any knowledge of their specific contents. OSCAR is designed to run even sophisticated applications in a 'cheap' environment (PCs + windows3.11 or microcontrollers). the lack of realtime support and efficient multitasking mechanisms in DOS made us to implement it in OSCAR;currently a priority driven multitasking management with soft and hard realtime capabilities, event handling and distributed execution on several machines is provided. Currently, OSCAR is running on two mobile platforms with different hardware structures and kinematics to proove it's functionality.
暂无评论