Abstract: With the continuous development of cars in the field of intelligence, intelligent parking technology has become the current research hotspot, and the detection of parking space is the premise of intelligent parking. Therefore, this paper proposes a parking space detection method based on machine vision. The method first carries out filtering processing, binarization processing, edge detection and other pretreatment work on the collected image, and then realizes the detection of parking space using Hough transform on the basis of the pretreatment. In the process of detection, this paper uses MATLAB software to implement the above algorithm, and builds a visual interactive interface for parking space detection based on the GUI platform. Experimental results show that the proposed method can accurately detect parking spaces, but it is also found in the process of the experiment, when the parking space area is seriously polluted, the detection effect of the detection algorithm will be reduced.

Key words: Machine vision; Parking space detection; Image processing; GUI

摘要： 随着汽车在智能化领域的不断发展，智能泊车技术已成为当前的研究热点，而停车位 的检测又是进行智能泊车的前提。据此，文章提出了一种基于机器视觉的停车位检测方法， 该方法首先对采集的图像进行滤波处理、二值化处理、边缘检测等预处理工作，随后在预处 理的基础上利用霍夫变换实现对停车位的检测。在检测的过程中，使用 MATLAB 软件对上述 算法进行实现，并基于 GUI 平台搭建了用于停车位检测的可视化交互界面。实验结果表明， 文章所提出的方法能够较为准确地实现对停车位的检测，但在实验过程中也发现，当停车位 区域污损较为严重时，该检测算法检测效果会降低。

关键词: 机器视觉；停车位检测；图像处理；GUI