Application of Optimization Algorithms to Multiple Kinect Integration in Construction of Operation Posture Hazard Assessment System

• Main Authors: Chin-Hao Chang, 張晉豪
• Other Authors: Fang-Chih Tien
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/a52hhe

碩士 === 國立臺北科技大學 === 工業工程與管理系碩士班 === 102 === Musculoskeletal injuries caused by Long-term poor posture, excessive force, highly repetitive work are common in all kinds of labor among industries. In order to improve occupational injuries caused by poor posture, working posture must be assessed by a variety of tools. However, the traditional methods of assessment are in recording and direct observation method based, will spend a lot of manpower and time. Capturing the human joints coordinates by Kinect, and combining with operation assessment tool to develop an automated posture operation hazard assessment system can improve the problem of the traditional methods which manpower and time consuming. However, due to the limited Kinect shooting range, and obscuring human joints by body or foreign objects, causing Kinect coordinate information can’t capture joints precisely, resulting in automated posture operation hazard assessment system error. The purpose of this study was to use multiple Kinect to shoot in different locations and using Optimization Algorithms to convert the coordinate of multiple kinect to the same coordinate system, and finally Integrate the skeleton information which capture by multiple Kinect, to facilitate the analysis and evaluation of various operating hazards pose risks.