Object detection and distance measurement with computer vision

• Main Author: Zhu, Jun
• Other Authors: Song Qing
• Format: Theses and Dissertations
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/76340
• Institution: Nanyang Technological University

With the development of computer and robotic technology, it becomes more and more common for robots to do hard, regular work for human. The appearance of robots has already changed humanity’s daily life and improved the efficiency of industry production. Among different kinds of robot applications, object detection and distance measurement are especially important for industry production. Like quality examination, product sorting and dimension measurement. These robots in these applications are programmable, high speed and high efficiency. To do object detection and distance measurement, deep learning and time of flight camera are widely used in industry. Engineers have already spent a lot of time doing related research. However, it is not easy and cheap to implement these methods. Deep learning needs so much computing power and industrial cameras are very expensive. Hence, in this report, the aim is to explore some easy ways to do object detection and distance measurement for some easy application in single background. To achieve this goal, some external sensors are necessary. Later the report will explain the theoretical basis and the experimental results of these two different ways. Then there will be a short discussion about the future development, which can improve the efficiency of these two methods covered in the report.