Moroccan Intelligent Transport System is the first Moroccan system that uses the latest advances in computer vision, machine learning and deep learning techniques to manage Moroccan traffic and road violations. In this paper, we propose a fully automatic approach to Multiple Hypothesis Detection and Tracking (MHDT) for video traffic surveillance. The proposed framework combines Kalman filter and data association-based tracking methods using YOLO detection approach, to robustly track vehicles in complex traffic surveillance scenes. Experimental results demonstrate that the proposed approach is robust to detect and track the trajectory of the vehicles in different situations such as scale variation, stopped vehicles, rotation, varying illumination and occlusion. The proposed approach shows a competitive results (detection: 94.10% accuracy, tracking: 92.50% accuracy ) compared to the state-of-the-art approaches.

Vehicle type classification is a critical functionality in any Intelligent Transportation System (ITS). In this paper, we present a novel two-layers vehicle type classification framework based on the vehicle’s 3D parameters and its local features. This framework is a part of the first Moroccan Video Intelligent Transport System (MOVITS) which aims to control traffic and road code violations. In the first layer, the 3D features are extracted using the disparity map generated from stereo-images, then the width, the height, and the length of the vehicle are calculated based on the obtained list of 3D points. In the second layer, a gradient-based method is applied to extract 2D features and a dimensionally reduction algorithm is performed to reduce its size. Both features are combined to construct the final feature vector which is used as an input for the classification. A Moroccan dataset and the BIT-dataset were used to, respectively, validate the proposed framework and conduct a comparative study with the state of the art algorithms. The experimental results demonstrate the efficiency of our approach against existing algorithms.

Automated License Plate Detection and Recognition (ALPR) is a technology that identifies and reads license plates in transportation and surveillance systems. However, ALPR is a challenging task due to several factors, including differences in vehicle license plates between countries, a lack of public datasets, environmental factors that can decrease the performance of recognition models, and challenging lighting conditions for vehicle license plates. In this work, we present an ALPR system for real-time applications specifically designed to detect and recognize Moroccan license plates. Our contribution is threefold. First, we employ the YOLOv3 algorithm to extract valuable features from input scenes and detect vehicles. Second, we track vehicles and assign a unique id to each detected vehicle. Finally, we detect and recognize the plates of tracked vehicles based on computer vision techniques and a voting system. The latter is a post-processing step that selects the most consistent content as the final prediction. Furthermore, we collected and annotated a set of videos that contain vehicles with Moroccan license plates in various conditions. The collected data is used to assess the performance of our system. The obtained results demonstrate that our system achieves an accuracy of 99.19% and 95.62% for detection and recognition, respectively.

In this paper, we propose a hybrid method for background modeling, subtracting, and extracting moving objects, which is based on the use of W⁴ and Extended Centre Symmetric Local Binary Pattern (XCS-LBP) approaches. Initially, we apply W4 to get the foreground mask of the video scene, after that we use XCS-LBP to clarify the results obtained. The main focus of this paper is to illustrate moving objects. In the evaluation section, we will prove that the proposed hybrid method gives better results than W⁴ with the use of datasets contains a different background.