Jiandan Chen

Planning of a Multi Stereo Visual Sensor System - Depth Accuracy and Variable Baseline Approach

This paper presents a method for planning the position of multiple stereo sensors in an indoor environment. This is a component in an intelligent vision agent system. We propose a new approach to dynamically adjust the multiple stereo pairs position, pose and baseline length in 3D space in order to get sufficient visibility and enough accuracy for surveillance, tracking and 3D reconstruction. The paper proposes visibility constraints to plan the cameras pose, and a depth accuracy constraint to control the baseline length. The minimum number of stereo pairs necessary to cover the target space is optimized by an integer linear programming. The 3D simulations of reconstruction accuracy and the human activities space coverage problem were performed in Matlab.

Arrangement of a Multi Stereo Visual Sensor System for a Human Activities Space

The first part of this chapter introduces a mathematical geometry model which is used to analyze the iso-disparity surface. This model can be used to dynamically adjust the positions, poses and baseline lengths of multiple stereo pairs of cameras in 3D space in order to get sufficient visibility and accuracy for surveillance, tracking and 3D reconstruction. The depth reconstruction accuracy is quantitatively analyzed by the proposed model. The proposed iso-disparity mathematical model presents possibility of reliable control of the iso-disparity curves’ shapes and intervals by applying the systems configuration and target properties. In the second part of this chapter, the key factors affecting the accuracy of 3D reconstruction are analysed. It shows that the convergence angle and target distance influence the depth reconstruction accuracy most significantly. The depth accuracy constraints are implemented in the model to control the stereo pair’s baseline length, position and pose. It guarantees a certain accuracy in the 3D reconstruction. The reconstruction accuracy is verified by a cubic reconstruction method. The optimization is implemented by applying the camera, object and stereo pair constraints into the integer linear programming.

An Adaptive Model of the Fuzzy Variable - Quality Index

In this article we discuss a model of quality as the fuzzy variable in order to better understand the concept and thus enable the further development of the variable. We propose a general method to estimate the quality index which can handle both qualitative and quantitative factors. The method uses a fuzzy neural network since the system learns how to integrate the human judgment of quality into a quantitative index. In our case study we have examined the measurement of image quality and pulp quality.

RFID Multi-Target Tracking Using the Probability Hypothesis Density Algorithm for a Health Care Application

The intelligent multi-sensor system is a system for target detection, identification and information processing for human activities surveillance and ambient assisted living. This paper describes RFID multi-target tracking using the Gaussian Mixture Probability Hypothesis Density, GM-PHD, algorithm. The multi target tracking ability of the proposed solution is demonstrated in a simulation and real environment. A performance comparison of the Levenberg-Marquardt algorithm with and without the GM-PHD filter shows that the GM-PHD algorithm improves the accuracy of tracking and target position estimation significantly. This improvement is demonstrated by a simulation and by a physical experiment.

Depth reconstruction uncertainty analysis and improvement - The dithering approach

The depth spatial quantization uncertainty is one of the factors which influence the depth reconstruction accuracy caused by a discrete sensor. This paper discusses the quantization uncertainty distribution, introduces a mathematical model of the uncertainty interval range, and analyzes the movements of the sensors in an Intelligent Vision Agent System. Such a system makes use of multiple sensors which control the deployment and autonomous servo of the system. This paper proposes a dithering algorithm which reduces the depth reconstruction uncertainty. The algorithm assures high accuracy from a few images taken by low-resolution sensors. The dither signal is estimated and then generated through an analysis of the iso-disparity planes. The signal allows for control of the camera movement. The proposed approach is validated and compared with a direct triangulation method. The simulation results are reported in terms of depth reconstruction error statistics. The physical experiment shows that the dithering method reduces the depth reconstruction error.

An Adaptive Quality Assessment System—Aspect of Human Factor and Measurement Uncertainty

In this paper, we discuss a model of quality that makes use of the fuzzily defined variable approach to better understand the concept and, thus, enables the further development of this variable. We propose a general method that may estimate a quality index (QI) that handles both qualitative and quantitative issues. The system further uses a neural network since the system learns how to integrate human factors into a quantitative QI. In our case study, we have examined the measurement of image quality and proposed a theoretical model of pulp quality.

Planning of a multiple sensor system for human activities space: aspects of iso-disparity surface

The Intelligent Vision Agent System, IVAS, is a system for automatic target detection, identification and information processing for use in human activities surveillance. This system consists of multiple sensors, and with control of their deployment and autonomous servo. Finding the optimal configuration for these sensors in order to capture the target objects and their environment to a required specification is a crucial problem. With a stereo pair of sensors, the 3D space can be discretized by an iso-disparity surface, and the depth reconstruction accuracy of the space is closely related to the iso-disparity curve positions. This paper presents a method to enable planning the position of these multiple stereo sensors in indoor environments. The proposed method is a mathematical geometry model, used to analyze the isodisparity surface. We will show that the distribution of the iso-disparity surface and the depth reconstruction accuracy are controllable by the parameters of such model. This model can be used to dynamically adjust the positions, poses and baselines lengths of multiple stereo pairs of cameras in 3D space in order to get sufficient visibility and accuracy for surveillance tracking and 3D reconstruction. We implement the model and present uncertainty maps of depth reconstruction calculated while varying the baseline length, focal length, stereo convergence angle and sensor pixel length. The results of these experiments show how the depth reconstruction uncertainty depends on stereo pairs baseline length, zooming and sensor physical properties.