Novel results in short-range visualization and vision systems based on gated imaging  
 

Professor Andrzej Sluzek
School of Computer Engineering,
concurrent appointment Deputy Director, Robotics Research Centre,
Nanyang Technological University
Singapore
E-mail: assluzek@ntu.edu.sg
 

Abstract: The idea of visibility enhancement using gated images has been known and applied for at least 20 years. The majority of gated imaging systems are used for enhancing capabilities of human vision under highly difficult conditions (fog, turbid water, strong ambient light, etc.). In most of the reported applications the range of gated-imaging systems is long enough comparing to the duration of the illumination pulses so that a simplified model of gated imaging can be used. Secondly, the systems capture individual images only and no attempts have been reported on more sophisticated visualization methods using gated images.

In this presentation, we focus on the above limitations of gated-imaging systems. First, we briefly discuss and explain the theory of short-range gated imaging. In vision-related applications, short-range gated imaging is needed only in very high turbidity conditions so that we focus on gated imaging in highly turbid media.

The presented ideas are illustrated by exemplary results of simulated and physical tests.

One of the conclusions is that in short-range applications of gated imaging the duration of the illumination pulses should be reduced so that the depth of visualization is very limited. We propose a technique for an automatic image fusion using a sequence of gated images captured by a precisely controlled pulsed laser. The fusion of several images can provide good quality of visualization over a wide range of depths. A feasibilty study system build using this principle will be briefly discussed. The system is controlled by an FPGA device and controls the gated imaging operation with the accuracy limited by a single clock cycle duration.

We believe that the proposed methods are a significant step towards system capable of providing visibility in extremely difficult scenarios (e.g. "inspecting areas of turbulent liquids, seeing through almost non-transparent liquids, etc.).

Short Biography of the Speaker:
Dr. Andrzej SLUZEK received his MSc, PhD and DSc degrees from the Warsaw University of Technology (Poland) where he worked until 1992. In 1987 he was a visiting fellow with the Indian Institute of Science (Bangalore, India). In 1991-92 he was a research fellow at the Univeristy of Birmingham (UK). In 1992 he joined the Nanyang Technological University (Singapore) where he is currently an associate professor in the School of Computer Engineering (Division of Computing Systems). As a concurrent appointment, he is a deputy director of the Robotics Research Centre of NTU (from 1994). His research interests include machine vision, digital systems, intelligent robotics and (recently) visual information retrieval digital systems. Dr. Sluzek published over 120 research paper (mostly in machine vision and robotics).