研究目的
To propose a method to identify parking spaces and obstacles based on visual sensor and laser device recognition methods to overcome the limitations of current automated parking technologies.
研究成果
The proposed method effectively identifies obstacles and parking spaces, including their sizes, even when the obstacles and background have similar colors. This approach offers a practical solution for automatic parking systems, with potential for further development in obstacle classification and parking space size acquisition.
研究不足
The method's effectiveness may be limited by the resolution of the camera and the precision of the laser grid. Future research could explore automatic classification of obstacles and more comprehensive experiments on parking space identification.
1：Experimental Design and Method Selection:
The method involves installing a laser transmitter on the car to produce a checkerboard-shaped laser grid on the ground, which changes shape when encountering obstacles. These changes are captured by a camera for image processing.
2：Sample Selection and Data Sources:
The experiment simulates various obstacles and scenes encountered in real-life parking environments using model cars.
3：List of Experimental Equipment and Materials:
Hardware includes a grid laser emitter and a camera. Software includes image processing and obstacle detection algorithms implemented in C++ using OpenCV library.
4：Experimental Procedures and Operational Workflow:
The laser emitter and camera are adjusted to the appropriate angle to render a checkerboard laser grid effect on the ground. Images of simulated scenes are collected, preprocessed, and analyzed for obstacle detection.
5：Data Analysis Methods:
Image processing techniques include gamma conversion image enhancement, grayscale processing, mean filtering, binarized edge detection, contour detection, and convex hull detection.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
