A real-time detection method for judging parking space occupancy

Abstract

The invention discloses a real-time detection method for judging the occupancy of a parking space, which belongs to the technical field of machine vision, and in particular relates to a real-time detection method for judging whether a parking space is correctly occupied by a detection method of combining a convolutional neural network with a depth map viewed from above. It specifically includes building a binocular stereo camera, collecting vehicle samples and non-vehicle samples, designing and training a vehicle detection model based on the convolutional neural network CarNet, manually calibrating the position of the monitoring parking space in the camera image, and obtaining the depth map of the monitoring scene 1. Segment the depth map to initially obtain the vehicle area, use the CarNet detection model to detect the vehicle, fuse the depth map and convolutional neural network results to obtain a reliable vehicle detection frame, and determine whether the parking space is occupied. The invention can be widely used in semi-open-air or indoor parking lots such as gas stations and underground parking lots, can accurately output the occupancy of parking spaces, has the advantages of low implementation cost, high work efficiency, strong automation ability, etc., and has broad development prospects.

Description

technical field [0001] The invention belongs to the technical field of machine vision, and in particular relates to a real-time detection method for judging parking space occupancy. Background technique [0002] In recent years, with the rapid development of the economy and the improvement of people's living standards, the per capita ownership of cars has also increased year by year. With the acceleration of the pace of automobiles entering the family, the resulting static traffic problems have become increasingly prominent, and have become prominent problems restricting the development of my country's automobile industry and even the urban national economy. The current problems are mainly manifested in the following aspects: first, the planning and layout are not reasonable; second, the control of land use indicators is insufficient; third, the number of parking spaces in the community is small; Tao phenomenon is serious. The above situation has seriously affected the dev...

AI Technical Summary

Problems solved by technology

However, the traditional parking lot is getting bigger and bigger, and the efficiency of manual management is getting lower and lower. Therefore, the intelligent parking lot has attracted more and more attention from the industry, and the research direction has been continuously broadened and deepened.

Images