Pallet Recognition and Localization Method for Vision Guided Forklift

For navigating automatic forklift, we proposed an method for locating of the forklift. To improve efficiently of image processing and robust of object identification, the color image is transformed from RGB space into HSV and YUV space. Afterwards, we find out the mid-point of the pallets, calculate position of pallets relative to forklift by camera space model which builds the relationship between image space and real world space. In order to improve system speed, we use the Kalman filter to decrease the processing data. Experiment results indicate that our method has good performance in improving accuracy and efficiency of localization.