A computer vision assisted geoinformation inventory for traffic infrastructure

Šegvić, Siniša; Brkić, Karla; Kalafatić, Zoran; Stanisavljević, Vladimir; Ševrović, Marko; Budimir, Damir; Dadić, Ivan

doi:10.1109/itsc.2010.5624979

Cited by 33 publications

(12 citation statements)

References 34 publications

(50 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Despite a large number of traffic-sign datasets, a comparison of traffic-sign detectors for large numbers of categories remains a challenging problem. In contrast to existing benchmarks that focus mostly on small numbers of supercategories (GTSDB [10]), or on small numbers of simple traffic signs (BTS [17], MASTIF [18], STSD [20], LISA [11]), our comprehensive dataset contains 200 traffic-sign categories, including a large number of categories with significant intracategory variability. The closest large-scale dataset is the Tsinghua-Tencent 100K dataset; however, their evaluation still focuses only on 45 simple traffic signs.…”

Section: Related Workmentioning

confidence: 99%

Deep Learning for Large-Scale Traffic-Sign Detection and Recognition

Tabernik

Skočaj

2020

IEEE Trans. Intell. Transport. Syst.

234

View full text Add to dashboard Cite

Automatic detection and recognition of traffic signs plays a crucial role in management of the traffic-sign inventory. It provides accurate and timely way to manage traffic-sign inventory with a minimal human effort. In the computer vision community the recognition and detection of traffic signs is a well-researched problem. A vast majority of existing approaches perform well on traffic signs needed for advanced driversassistance and autonomous systems. However, this represents a relatively small number of all traffic signs (around 50 categories out of several hundred) and performance on the remaining set of traffic signs, which are required to eliminate the manual labor in traffic-sign inventory management, remains an open question. In this paper, we address the issue of detecting and recognizing a large number of traffic-sign categories suitable for automating traffic-sign inventory management. We adopt a convolutional neural network (CNN) approach, the Mask R-CNN, to address the full pipeline of detection and recognition with automatic end-to-end learning. We propose several improvements that are evaluated on the detection of traffic signs and result in an improved overall performance. This approach is applied to detection of 200 traffic-sign categories represented in our novel dataset. Results are reported on highly challenging trafficsign categories that have not yet been considered in previous works. We provide comprehensive analysis of the deep learning method for the detection of traffic signs with large intra-category appearance variation and show below 3% error rates with the proposed approach, which is sufficient for deployment in practical applications of traffic-sign inventory management.Index Terms-Deep learning, Traffic-sign detection and recognition, Traffic-sign dataset, Mask R-CNN, Traffic-sign inventory management.

show abstract

Section: Related Workmentioning

confidence: 99%

Deep Learning for Large-Scale Traffic-Sign Detection and Recognition

Tabernik

Skočaj

2020

IEEE Trans. Intell. Transport. Syst.

234

View full text Add to dashboard Cite

show abstract

“…Em conformidade com Uddin e Akhi [10], o conjunto para treinamento deve ser escolhido de forma que não confunda o algoritmo de aprendizagem, desse modo, foram utilizadas ao todo 6194 imagens, onde: 2114 possuem somente as placas de velocidade máxima permitida, 1372 contém apenas placas de parada obrigatória e 2708 não possuem nenhum dos dois objetos de interesse. As imagens positivas são oriundas da internet e dos bancos de imagens: German Traffic Sign Recognition Benchmarck (GTSRB) [8], The MASTIF dataset [23] e BelgiumTS dataset [24]. As Fig.…”

Section: Classificadores Em Cascataunclassified

Algoritmo Haar Cascade Aplicado na Detecção das Placas de Parada Obrigatória e de Velocidade Máxima Permitida

Macedo¹,

Gomes²

2020

Anais Do XI Computer on the Beach - COTB '20

View full text Add to dashboard Cite

Systems able to assist drivers in the safe driving of vehiclesprovide several advantages, such as the reduction of trafficaccidents, mostly with fatalities, normally caused by humanfailures, whether for distractions or even problems related tolighting or climate change. Based on this, this research aims topresent a computational model capable of detect stop signs andspeed limit signs, so that it contributes to the development ofprogressively intelligent vehicles. The system was implementedin Phyton programming language, with the support of OpenCVlibrary, and it was divided into two steps: firstly it wasperformed the training and classification of the objects throughHaar Cascade classification method, and in the second step, inorder to improve the results, colors relevant to the object wereidentified using the HSV color space. During the experiments,the proposed algorithm presented satisfactory results, with a hitrate of 91% for speed limit signs and 93% for stop signs. In orderto refine the proposed solution, it is intended for the next stepsto include traffic sign information recognition, to either describethe specified speed on the detected objects and further reducefalse positives.

show abstract

“…TSR systems can be used for maintenance of traffic signs or roads. In [17] and [18], TSR systems were utilized to check the condition of traffic signs along the major roads. Wen et al [19] utilized mobile laser scanning data for spatial-related traffic sign inspection.…”

Section: B Machine Vision Based Tsr Systems and Their Applicationsmentioning

confidence: 99%

“…It is collected in different parts of Sweden. 9) MAPPING AND ASSESSING THE STATE OF TRAFFIC INFRASTRUCTURE (MASTIF) DATASET [18] This dataset consists of three small datasets collected in 2009, 2010 and 2011, respectively. The dataset-2009 is a classification dataset containing 6, 423 signs from 97 classes.…”

Section: ) Stereopolis Database [28]mentioning

confidence: 99%

Machine Vision Based Traffic Sign Detection Methods: Review, Analyses and Perspectives

Liu

Chang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Traffic signs recognition (TSR) is an important part of some advanced driver-assistance systems (ADASs) and auto driving systems (ADSs). As the first key step of TSR, traffic sign detection (TSD) is a challenging problem because of different types, small sizes, complex driving scenes, and occlusions. In recent years, there have been a large number of TSD algorithms based on machine vision and pattern recognition. In this paper, a comprehensive review of the literature on TSD is presented. We divide the reviewed detection methods into five main categories: color-based methods, shape-based methods, color-and shape-based methods, machine-learning-based methods, and LIDAR-based methods. The methods in each category are also classified into different subcategories for understanding and summarizing the mechanisms of different methods. For some reviewed methods that lack comparisons on public datasets, we reimplemented part of these methods for comparison. The experimental comparisons and analyses are presented on the reported performance and the performance of our reimplemented methods. Furthermore, future directions and recommendations of the TSD research are given to promote the development of the TSD.

show abstract

A computer vision assisted geoinformation inventory for traffic infrastructure

Cited by 33 publications

References 34 publications

Deep Learning for Large-Scale Traffic-Sign Detection and Recognition

Deep Learning for Large-Scale Traffic-Sign Detection and Recognition

Algoritmo Haar Cascade Aplicado na Detecção das Placas de Parada Obrigatória e de Velocidade Máxima Permitida

Machine Vision Based Traffic Sign Detection Methods: Review, Analyses and Perspectives

Contact Info

Product

Resources

About