Imaging objects behind a partially reflective surface with a modified time-of-flight sensor

Time-of-Flight (ToF) methods are used in different applications for depth measurements. There are mainly 2 types of ToF measurements, Pulsed Time-of-Flight and Continuous-Wave Time-of-Flight. Pulsed Time-of-Flight (PToF) techniques are mostly used in combination with a scanning mirror, which makes them not well suited for imaging purposes. Continuous-wave Time-of-Flight (CWToF) techniques are mostly used wide-field, hence they are much faster and more suited for imaging purposes but cannot be used behind partially-reflective surfaces. In commercial applications, both ToF methods require specific hardware, which cannot be exchanged. In this paper, we discuss the transformation of a CWToF sensor to a PToF camera, which is able to make images and measure the distances of objects behind a partially-reflective surface, like the air-water interface in swimming pools when looking from above. We first created our own depth camera which is suitable for both CWToF and PToF. We describe the necessary hardware components for a normal ToF camera and compare it with the adapted components which make it a range-gating depth imager. Afterwards, we modeled the distances and images of one or more objects positioned behind a partially-reflective surface and combine it with measurement data of the optical pulse. A scene was virtualized and the rays from a raytracing software tool were exported to Matlab(TM). Subsequently, pulse deformations were calculated for every pixel, which resulted in the calculation of the depth information.

Publication year:2014

Keywords:3d cameras, Time-of-flight, imaging systems, depth sensing, 3d imaging, range finding