A Computer Vision-Based Roadside Occupation Surveillance System for Intelligent Transport in Smart Cities

Abstract

In digital and green city initiatives, smart mobility is a key aspect of developing smart cities and it is important for built-up areas worldwide. Double-parking and busy roadside activities such as frequent loading and unloading of trucks, have a negative impact on traffic situations, especially in cities with high transportation density. Hence, a real-time internet of things (IoT)-based system for surveillance of roadside loading and unloading bays is needed. In this paper, a fully integrated solution is developed by equipping high-definition smart cameras with wireless communication for traffic surveillance. Henceforth, this system is referred to as a computer vision-based roadside occupation surveillance system (CVROSS). Through a vision-based network, real-time roadside traffic images, such as images of loading or unloading activities, are captured automatically. By making use of the collected data, decision support on roadside occupancy and vacancy can be evaluated by means of fuzzy logic and visualized for users, thus enhancing the transparency of roadside activities. The CVROSS was designed and tested in Hong Kong to validate the accuracy of parking-gap estimation and system performance, aiming at facilitating traffic and fleet management for smart mobility.

Deep Learning Model Comparison for Vision-Based Classification of Full/Empty-Load Trucks in Earthmoving Operations

“…Apart from vehicle classification or detection, CNN is also able to recognize the working or idle state of earthwork machines, like excavators or trucks [45], and CNN-TL can benefit earthmoving operations or related construction management [27,46]. Other non-CNN machine vision methods also have applications in related areas, like vehicle collision prediction or construction machine detection, [47][48][49][50]. It can be seen that current vision-based deep learning researches mainly focus on the vehicle classification or state identification of earthwork machinery, and CNN and CNN-TL are widely applied.…”

Vision-Based Potential Pedestrian Risk Analysis on Unsignalized Crosswalk Using Data Mining Techniques

“…Around the world, many cities have adopted information and communication technologies (ICT) to create intelligent platforms within a broader smart city context, and use data to support the safety, health, and welfare of the average urban resident [1,2]. However, despite the proliferation of technological advancements, road traffic accidents remain a leading cause of premature deaths, and rank among the most pressing transportation concerns around the world [3,4].…”

“…In general, there are two ways to support road users' safety; (1) passive safety systems such as speed cameras and fences which prevent drivers and pedestrians from engaging in risky or illegal behaviors; and (2) active safety systems which analyze historical accident data and forecast future driving states, based on vehicle dynamics and specific traffic infrastructures. A variety of studies have reported on examples of active safety systems, which include (1) the analysis of urban road infrastructure deficiencies and their relation to pedestrian accidents [6]; and (2) using long-term accident statistics to model the high fatality or injury rates of pedestrians at unsignalized crosswalks [7,8].…”

“…The novel contributions of this study are: (1) Repurposing video footage from CCTV cameras to contribute to the study of unsafe road conditions for pedestrians; (2) automatically extracting behavioral features from a large dataset of vehicle-pedestrian interactions; and (3) applying statistics and machine learning to characterize and cluster different types of vehicle-pedestrian interactions, to identify those with the highest potential risk to pedestrians. To the best of our knowledge, this is the first study of potential pedestrian risky event analysis which creates one sequential process for detecting and segmenting objects, extracting their features, and applying data mining to classify PPREs.…”

Video-Based Parking Occupancy Detection for Smart Control System

“…Computer vision-based methods reduce costs and display real-time sequence images to observe multiple objects. Studies [17,18] have proposed image-based vehicle detection methods, image fragmentation into gray levels, and segment area distribution analysis to detect the presence of vehicles. However, traditional image processing has difficulties with processing complex backgrounds such as streets, bright or dark targets, and large areas with occlusion, as well as with separating targets and backgrounds.Several machine learning algorithms and object classification methods have been developed over many years.…”