25-08-2017, 09:32 PM
Human Motion Regeneration using Sensors
Human motion analysis is receiving increasing attention from computer vision researchers. This interest is motivated by a wide spectrum of applications, such as athletic performance analysis, surveillance, man-machine interface, content-based image storage and retrieval, and video conferencing. Three major areas related to interpreting human motion are motion analysis involving human body parts, tracking of human motion using single or multiple cameras, and recognizing human activities from image sequences. Motion analysis of human body parts involves the low level segmentation of the human body into segments connected by joints, and regeneration of the 3D structure of the human body using its 2D projections over a sequence of images. Tracking human motion using a single camera or multiple cameras focuses on higher-level processing, in which moving humans are tracked without identifying specific parts of the body structure. Understanding the human movements or activities based on the moving human image naturally comes after successfully matching the moving human image from one frame to another in image sequences. We will look at athletic performance analysis and regeneration technique in this write up. The process of analyzing the gait or motion of subject is often referred to as motion analysis. Motion analysis, in this context, is the detailed study of human motion, in a certain task or within a certain area. Motion analysis methods are mostly dependent on vision systems. Vision system methods may not adjust and accommodate easily to major changes made in the environment or lighting condition of the area. These methods are time consuming and as such do not suit sports performance analysis. Therefore, this approach is often not practical in a sports monitoring/training situation. For such situations, sensor based motion analysis is used. This is very much useful to coaches as they obtain required results fast. Thus sensor based motion analysis helps sports performance enhancement and prevention of injury. Accelerometers play an important role in short term supervised monitoring and long term unsupervised monitoring .Tri-axial accelerometers provide information on the acceleration in three directions, namely the vertical (Z-axis), anterior and posterior (Y-axis)directions and lateral direction(X-axis The accelerometers are placed on the test subjectâ„¢s body at points of interest, to measure the acceleration at those points. These are then strapped down to the test subjectâ„¢s body using elastic body straps. The accelerometers have to be securely strapped down to the test subjectâ„¢s body to ensure that the measurement obtained is purely due to the movement of the test subject and not due to the movement of the accelerometers within the body straps. Placement of accelerometers on the test subjectâ„¢s body Sensors establish communication with a PC via a USB base station, using very high frequency part of em spectrum of the order of 900 MHz as its communication medium. The USB base station attached to a computer enables end user to issue various commands to the accelerometers. Data got from accelerometers is then conditioned and converted into appropriate kinematical information, which then gets displayed in numerical or graphical form. An important feature of the system is that initial processes, data acquisition and data interpretation run on a single programming platform. In order to display the results acquired from the accelerometer in terms of a regenerated motion, a stick figure is designed to function with the accelerometer control system. A stick figure is a simple type of drawing to depict the general form of humans, for displaying results of motion regeneration. All calculations for positions start from the tip of the right foot. Most drawings of the human body take the length of the head as the reference for calculating lengths of different parts of the body. The calculation for the stick figure involves simple trigonometric functions. The initial point of reference is taken to be at node 14(the right foot) and moved upwards from that point to
Human motion analysis is receiving increasing attention from computer vision researchers. This interest is motivated by a wide spectrum of applications, such as athletic performance analysis, surveillance, man-machine interface, content-based image storage and retrieval, and video conferencing. Three major areas related to interpreting human motion are motion analysis involving human body parts, tracking of human motion using single or multiple cameras, and recognizing human activities from image sequences. Motion analysis of human body parts involves the low level segmentation of the human body into segments connected by joints, and regeneration of the 3D structure of the human body using its 2D projections over a sequence of images. Tracking human motion using a single camera or multiple cameras focuses on higher-level processing, in which moving humans are tracked without identifying specific parts of the body structure. Understanding the human movements or activities based on the moving human image naturally comes after successfully matching the moving human image from one frame to another in image sequences. We will look at athletic performance analysis and regeneration technique in this write up. The process of analyzing the gait or motion of subject is often referred to as motion analysis. Motion analysis, in this context, is the detailed study of human motion, in a certain task or within a certain area. Motion analysis methods are mostly dependent on vision systems. Vision system methods may not adjust and accommodate easily to major changes made in the environment or lighting condition of the area. These methods are time consuming and as such do not suit sports performance analysis. Therefore, this approach is often not practical in a sports monitoring/training situation. For such situations, sensor based motion analysis is used. This is very much useful to coaches as they obtain required results fast. Thus sensor based motion analysis helps sports performance enhancement and prevention of injury. Accelerometers play an important role in short term supervised monitoring and long term unsupervised monitoring .Tri-axial accelerometers provide information on the acceleration in three directions, namely the vertical (Z-axis), anterior and posterior (Y-axis)directions and lateral direction(X-axis The accelerometers are placed on the test subjectâ„¢s body at points of interest, to measure the acceleration at those points. These are then strapped down to the test subjectâ„¢s body using elastic body straps. The accelerometers have to be securely strapped down to the test subjectâ„¢s body to ensure that the measurement obtained is purely due to the movement of the test subject and not due to the movement of the accelerometers within the body straps. Placement of accelerometers on the test subjectâ„¢s body Sensors establish communication with a PC via a USB base station, using very high frequency part of em spectrum of the order of 900 MHz as its communication medium. The USB base station attached to a computer enables end user to issue various commands to the accelerometers. Data got from accelerometers is then conditioned and converted into appropriate kinematical information, which then gets displayed in numerical or graphical form. An important feature of the system is that initial processes, data acquisition and data interpretation run on a single programming platform. In order to display the results acquired from the accelerometer in terms of a regenerated motion, a stick figure is designed to function with the accelerometer control system. A stick figure is a simple type of drawing to depict the general form of humans, for displaying results of motion regeneration. All calculations for positions start from the tip of the right foot. Most drawings of the human body take the length of the head as the reference for calculating lengths of different parts of the body. The calculation for the stick figure involves simple trigonometric functions. The initial point of reference is taken to be at node 14(the right foot) and moved upwards from that point to