A solution to determine the positions of rntransportation vehicles without any optical markers rnin an area under camera rnsurveillance was introduced. This solution was based rnon a net of cameras on the ceiling of the vehicle rnoperational areas. A robust rnbackground subtraction method was performed to detect rnthe contours of all moving objects in every camera rnframe. These so called rnforeground objects contours were fed into a particle rnfilter to estimate the position, orientation and rnspeed of the detected objects over a rncertain number of frames. Based on the camera data rnthis so-called state estimation was performed for all rnforeground objects like rnhumans, other moving obstacles and the vehicles. Due rnto the fact that the vehicles can be totally covered rnby their payload, they rncannot be detected by color or shape. Instead, the rnvehicles were identified by comparing their movement rnwith the movement of all rndetected foreground objects.
展开▼
机译:Nívelderuídonoambiente de trabalho do教授deducaçãofísicoeaulas de ciclismo indoor Nivel de ruido en el ambiente de trabajo del profesordeducaciónfísiconaulas de ciclismo indoor室内自行车课程中体育教师的工作环境噪音水平