Automated Real-Time Roadway Asset Inventory using Artificial Intelligence

Abstract

Roadway asset inventory data are essential in making data-driven asset management decisions. Despite significant advances in automated data processing, the current state of the practice is semi-automated. This paper demonstrates integration of the state-of-the-art artificial intelligence technologies within a practical framework for automated real-time identification of traffic signs from roadway images. The framework deploys one of the very latest machine learning algorithms on a cutting-edge plug-and-play device for superior effectiveness, efficiency, and reliability. The proposed platform provides an offline system onboard the survey vehicle, that runs a lightweight and speedy deep neural network on each collected roadway image and identifies traffic signs in real-time. Integration of these advanced technologies minimizes the need for subjective and time-consuming human interventions, thereby enhancing the repeatability and cost-effectiveness of the asset inventory process. The proposed framework is demonstrated using a real-world image dataset. Appropriate pre-processing techniques were employed to alleviate limitations in the training dataset. A deep learning algorithm was trained for detection, classification, and localization of traffic signs from roadway imagery. The success metrics based on this demonstration indicate that the algorithm was effective in identifying traffic signs with high accuracy on a test dataset that was not used for model development. Additionally, the algorithm exhibited this high accuracy consistently among the different considered sign categories. Moreover, the algorithm was repeatable among multiple runs and reproducible across different locations. Above all, the real-time processing capability of the proposed solution reduces the time between data collection and delivery, which enhances the data-driven decision-making process.

Application of artificial intelligence in road network inventory and network-wide road safety assessment

“…This technique was first used in the early 2000s, and it has since been refined to the point where it can now be used to identify a wide variety of road defects, including cracks, potholes, and missing lane markings. [11] In recent years, there has been a growing interest in using AI for more complex tasks related to road asset inventory, such as pavement condition assessment and traffic flow analysis. These tasks require a deeper understanding of road conditions and traffic patterns, and they can be more challenging to automate.…”

“…A recent proposed framework [11] improves the repeatability and cost-effectiveness of the asset inventory process by integrating advanced artificial intelligence technologies for automated real-time detection, classification, and localization of traffic signs from roadway images. This reduces the need for subjective and time-consuming human interventions.…”

Mobil haritalama amaçlı Mobilenet tabanlı trafik işaretleri tespit sistemi: kitlesel coğrafi bilgi toplama sistemi

“…Model, işaretlerin varlığına yeterince duyarlı değilse, o zaman daha fazla false negative değerine sahip olacaktır ve dolayısıyla recall daha yüksek bir değerde olacaktır. Presicion ve recall arasındaki bu potansiyel dengesizlik senaryoları nedeniyle, F1 Score gözetilerek, presicion ve recall arasında bir denge aranması tavsiye edilir[28,37]. Sistem Mimarisi (System Architecture) (predicted bounding box) ile gerçek sınırlayıcı kutunun kesişim alanı ile birleşim alanının oranıdır (Şekil 4).…”

“…Çalışmada, 0,50 ve üzeri IoU değerine sahip örnekler kullanılmıştır. Bu, yalnızca tahmini sınırlayıcı kutu ile gerçek sınırlayıcı kutu arasında %50'den fazla örtüşme olan tespitlerin dikkate alınacağı anlamına gelir[28,37].Önerilen bütünleşmiş metodolojinin sonuçları, IoU, mAP, recall ve F1 Score metriklerine göre ele alınmıştır. Şekil 5, test veri setinde toplanan görüntülerde saptanan işaretlerin örneklerini göstermektedir.…”