All Pixels Calibration for ToF Camera

Abstract

ToF (Time of Flight) camera is a new three-dimensional measurement technology. ToF camera can acquire depth images at high frame rates. In this paper, the influence of different measurement parameters on the accuracy of camera depth data is studied, the result shows that Integration time and measurement distance have great influence on the accuracy of camera depth data. While the quality of the pixel's depth data is closely related to the pixel's amplitude data. Therefore an integration time auto adaptation method based on amplitude data is proposed, which makes each pixel obtain the depth information under the best conditions. The experimental results show that ToF camera can effectively capture depth data of the complex scenes with distant and close shots by applying the integration time adaptation method. Finally, Gaussian process regression (GPR) model is used to calibrate the range errors of ToF camera. Experimental results show that this method can effectively calibrate the whole pixel.