A Real-time Handheld 3D Temperature Field Reconstruction System

Abstract

This paper presents a real-time and handheld 3D temperature field SLAM system which is not only useful in search and rescue robots, but also in building energy auditing, and fault diagnose of critical equipments in factories. The 3D temperature field SLAM system is built on a thermal camera and an RGB-D camera. The problem is how to fuse the temperature information with the RGB information to estimate pose of the camera and realize the 3D temperature field reconstruction in real-time. To solve this problem, we firstly use a simple method to fuse the thermal temperature with RGB-D information to get RGB-DT images. After that, we use RGB-DT images to estimate the sensor pose by fusing a depth-ICP algorithm and a direct RGB-DT method. The proposed handheld 3D temperature field SLAM system is advantageous due to its mechanical simplicity, mobility, low weight and real-time reconstruction ability. We carried out several experiments to demonstrate the performance of the proposed system. The experiment results show that our system can achieve large-scale 3D temperature field reconstruction in real-time.