Search PubMedSearch

Biomedical subjects

F Gazzani

Publications and source records attributed to F Gazzani.

3 recordsLinked to original sources

Performance of a 7-parameter DLT method for the calibration of stereophotogrammetric systems using 1-D transducers.

A method is presented for calibrating stereophotogrammetric systems which are based on 1-D cameras, i.e., able to record only one of the two coordinates relative to the projection of the target point onto the image plane of a camera. This procedure is based on the Direct Linear Transformation, DLT, introduced by Marzan and Karara. The present version (7DLT) is different from the standard 11-parameter DLT (11DLT); it avoids certain non-orthogonality problems arising in usual DLT implementations, whose correction can be obtained by modifying the basic algorithm, at the expense of its simplicity (MDLT). The accuracy assessment of the 7DLT method was carried out both by a computer simulation of a two-camera stereophotogrammetric system, and on real data from three 1-D optoelectronic cameras. The computer simulation permitted us to compare the performance of 7DLT with 11DLT and MDLT. The reconstruction accuracies of the three algorithms were quite similar. When using real 1-D data, a reconstruction accuracy of 0.66 mm RMS error was obtained (mean value for X, Y, Z coordinates) for 18 target points used as calibration control points. The extrapolation accuracy was 1.85 mm RMS for 180 target points lying outside the calibration structure.

Animals

Lumbar spine loading during half-squat exercises.

Evaluation of the compressive load acting on the lumbar spine (L3-L4) during half-squat exercises executed with a barbell resting on the subject's shoulders was undertaken. The kinematics of the upper body segments of two male and two female subjects as well as the barbell were described using data obtained by means of an optoelectronic system (CoSTEL). L3-L4 compressive load was calculated using a model of the anatomy of the trunk musculoskeletal system. Filtered surface electromyographic trunk flexor recordings from the obliquus externus and rectus abdominis and trunk extensor erectores spinae muscles as well as measurement of the ground reaction forces were also carried out for predicted result validation. During half-squat exercises with barbell loads in the range 0.8 to 1.6 times body weight the compressive loads on the L3-L4 segment vary between 6 and 10 times body weight. Erectores spinae contraction force was predicted to be between 30 and 50% of the relevant maximal isometric force. The magnitude of trunk flexion was found to be the variable which influenced most spinal compression load.

Adult