Wolfgang Weihs

Movement Artefacts in Range Images of Time-of-Flight Cameras

Modern time-of-flight measuring camera systems are able to provide accurate depth images but there are problems with regard to their usage in a dynamic environment due to their measurement principle. The Centre for Sensor Systems recently proposed a new 2D/3D camera avoiding this drawback by the monocular combination of a 2D and a 3D sensor. In this paper, the camera is presented and an approach for detecting erroneous depth values due to movement is made.

Detection and Classification of Moving Objects-Stereo or Time-of-Flight Images

This paper describes a system for detection and classification of moving objects based on support vector machines (SVM) and using 3D data. Two kinds of camera systems are used to provide the classification system with 3D range images: time-of-flight (TOF) camera and stereo vision system. While the former uses the modulated infrared lighting source to provide the range information in each pixel of a photonic mixer device (PMD) sensor, the latter employs the disparity map from stereo images to calculate three dimensional data. The proposed detection and classification system is used to classify different 3D moving objects in a dynamic environment under varying lighting conditions. The images of each camera are first preprocessed and then two different approaches are applied to extract their features. The first approach is a computer generated method which uses the principal component analysis (PCA) to get the most relevant projection of the data over the eigenvectors and the second approach is a human generated method which extracts the features based on some heuristic techniques. Two training data sets are derived from each image set based on heuristic and PCA features to train a multi class SVM classifier. The experimental results show that the proposed classifier based on range data from TOF camera is superior to that from the stereo system

Development and investigation of a long-range time-of-flight and color imaging system.

Time-of-flight (Tof) imaging based on the photonic mixer device (PMD) or similar ToF imaging solutions has been limited to short distances in the past, due to limited lighting devices and low sensitivity of ToF imaging chips. Long-range distance measurements are typically the domain of laser scanning systems. In this paper, PMD based medium- and long-range lighting devices working together with a 2-D/3-D camera are presented and several measurement results are discussed. The proposed imaging systems suffer from two systematic limitations in addition to problems due to wind and insufficient lighting: a low lateral resolution of the depth imaging chip and ambiguities in the distance measurements. In order to provide a robust and flexible system, we introduce algorithms to obtain unambiguous depth values (phase unwrapping) and to perform a joint motion compensation and super-resolution. Several experiments were conducted in order to evaluate the components of the multimodal imaging system.

2D/3D Image Data Analysis for Object Tracking and Classification

Object tracking and classification is of utmost importance for different kinds of applications in computer vision. In this chapter, we analyze 2D/3D image data to address solutions to some aspects of object tracking and classification. We conclude our work with a real time hand based robot control with promising results in a real time application, even under challenging varying lighting conditions.

Scanning 2D/3D monocular camera

We present the principal aspects and the concept of a monocular combination of a scanning 3D time-of-flight sensor with a large-scale conventional 2D image sensor. While the 2D sensor profits from the whole field of view of an F-Mount photo film format lens, the smaller-sized 3D sensor is mounted onto a highly precise XY linear move stage. Thus, by the means of macro-scanning, the 3D sensor can be moved to interesting parts of the scene so as to provide a second modality e.g. for classification purposes, or, in the same setup, micro-scanning can be applied to enhance the lateral 3D resolution. A test setup was realized to verify the performance.

Illumination, phase step optimization and improvements in simultaneous multiple frequency measurement for Time-of-Flight sensors

The systematic linearity error due to harmonics present in a Time-of-Flight (TOF) sensor system contributes to the inaccuracy of its three dimensional (3D) depth measurements. In this paper, a modified method which combines illumination waveform optimization and phase stepping algorithm selection is proposed to achieve a better performance at measurement linearity through numerical and practical experiment. Additionally, a solution of high-order harmonic reduction in simultaneous multiple frequency framework is investigated for unambiguous range and acceptable high frame rate.

Range Image Based Classification System Using Support Vector Machines

This paper describes a classification system based on Support Vector Machines (SVM) and using 3D range images. Two kinds of camera systems are used to provide the classification system with 3D range images: Time-oF-Flight (TOF) camera and Stereo Vision System. While the former uses the modulated infrared lighting source to provide the range information in each pixel of a Photonic Mixer Device(PMD) sensor, the latter employs the disparity map from stereo images to calculate three dimensional data. The proposed detection and classification system is used to classify different 3D moving objects in a dynamic environment under varying lighting conditions. The images of each camera are first preprocessed and then two different approaches are applied to extract their features. The first approach is a Computer Generated method which uses the Principal Component Analysis (PCA) to get the most relevant projection of the data over the eigenvectors and the second approach is a Human Generated method which extracts the features based on some heuristic techniques. Two training data sets are derived from each image set based on heuristic and PCA features to train a multi class SVM classifier. The experimental results show that the proposed classifier based on range data from TOF camera is superior to that from the stereo system.