Indian Journal of Chemical Technology Vol. 26, May 2019, pp. 274-278 Improving estimation of body lengths using extended Kalman Filter for squat movement Hüseyin Eski* ,1 , Ümit Kocabıçak 2 & Ertuğrul Gelen 3 1,2 Computer Engineering Department, Faculty of Computer and Information Sciences, Sakarya University, Sakarya, Turkey. 3 Education of Coaching Department, Faculty of Sport Sciences, Sakarya Applied Sciences University, Sakarya, Turkey. E-mail: heski@sakarya.edu.tr Received 20 February 2019; accepted 18 April 2019 Modeling and analyzing of human movements has become easier with the development of sensor technologies. Human movements can be modeled using image processing software with depth and motion sensors in 3D. Measurement errors are also observed in motion detection sensors as in most systems. Special filters have to be developed for each system in order to minimize this error rate and obtain more realistic measurements. Kalman Filter is a well-known method that is commonly used to minimize this type of measurement errors. In this study, the actual body lengths (upper arm, forearm, lower leg, upper leg) are measured and obtained from the human motion sensor. Kalman Filter and Extended Kalman Filter are applied to the obtained data from human motion sensor. All measurements are compared with the actual body lengths and error rate is calculated as using Mean Absolute Percentage Error (MAPE). Kinect data are compared with actual lengths and error rates were calculated at 20%, when the Kalman Filter is applied, the error rate decreased to 14%, while when the Extended Kalman filter is applied, it dropped to 8%. Human motion sensor data have been improved with using Extended Kalman Filter. Thus, actual measurements of candidatescan be easily obtained with only one useful sensor without taking any actual measurements by saving time and budget. Keywords: Kalman filter, Extended Kalman filter, Kinematic analysis, Squat movement, Human motion analysis With the rapid development of computer vision technology, human motion capture and analysis are widely used in many research areas, such as health, ergonomics, sports and security. Some information such as the location and position of person, and coordinates of joints should be known to analyze human movements. Such as marker usage, wearable inertial sensors and computer vision based algorithms are effectively used in the motion detection in recent studies. Marker usage 1-4 and wearable inertial sensors 4-7 are widely used in the digitization of limited inland and simple movements, especially in the film industry, increased reality and simulation applications. It is known that wearable inertial sensors have limited mobility and are not suitable for large areas. In the case of marker usage, markers can not be detected from time to time and measurement is not possible. In addition to these, the cost of using multiple cameras is very high. When image processing management is used from the videos; the person is asked to mark the joint points individually by the user, resulting in workload and time loss. Furthermore, movement analysis of disabled people with the help of these technologies and physical therapy methods are being developed 5,8,9 . For these systems, specially equipped spaces and high costs are required. It can also be installed only on certain health campuses due to the cost and low portability of the equipment used. In terms of sustainability as well as in the installation phase, there is a need for specialized human power in this area. The fact that the technological infrastructure costs are high, not easily accessible and portable, and that there are few specialists in this area, this field has become a necessity for new studies. In the study of Guimarães et al., ergonomic risk analysis was performed in 30 chemistry laboratories and working conditions were analyzed. As a result of the analysis, a number of ergonomic problems have been identified and they have developed a 3D virtual simulation tool for their solution. There ergonomic problems determined as repetitive and static awkward postures as squat movement, trunk and neck forward bending, shoulder flexion and abduction over 90 degrees; manual material handling activities, as lifting, carrying, pushing etc. 10