Research on Parking Space Status Recognition Method Based on Computer Vision

To improve the utilization rate of parking space resources and reduce cost installing maintaining sensor recognition, this paper proposed an improved computer vision-based status recognition method. The overall accuracy was by graying video, filtering smoothing noise reduction, image enhancement pre-processing, introducing texture feature extraction method based on LBP operator, improving background difference method, then, we used a perceptual hash algorithm to calculate Hamming distance between string current frame excluding influence light pedestrian accuracy. Finally, system is developed relied Python environment, spaces are recognized in three environmental states: with direct light, without rain snow. average experimental results 97.2%, which verifies model.