This article presents a comparison of two artificial intelligence techniques used for automatic and continuous vehicular parking control. The techniques employed are Artificial Vision and Convolutional Neural Networks (CNN). The first technique is based on the analysis of images and videos of parking lots using the OpenCV library for computer vision and the Python programming language. It involves identifying and defining the coordinates of the region of interest (ROI) for parking spaces. Using artificial vision, the detection of available or occupied parking spaces is carried out using the Gaussian blur technique, followed by converting the images from RGB to grayscale. Subsequently, the original ROIs and the converted image's ROI are evaluated, and the standard deviation and average are calculated to check if they are above or below a threshold. The second technique utilizes object detectors based on CNN, such as the YOLO (You Only Look Once) architecture, specifically YOLO V7. Each technique was evaluated through tests, yielding an accuracy of 0.88 and 0.82 for precision and sensitivity, respectively, with Convolutional Neural Networks. The results obtained with artificial vision were 0.80 and 0.79 for precision and sensitivity, respectively. Based on the obtained results, better outcomes were achieved with the use of Convolutional Neural Networks.