Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 40 of the 41 references it cites were found in a registry or an index.

Dynamic Crosswalk Scene Understanding for the Visually Impaired

Open access
Tian, Shishun, Zheng, Minghuo, Zou, Wenbin, Li, Xia, Zhang, Lu · 2021 · IEEE Transactions on Neural Systems and Rehabilitation Engineering

Abstract

Independent mobility poses a great challenge to the visually impaired individuals. This paper proposes a novel system to understand dynamic crosswalk scenes, which detects the key objects, such as crosswalk, vehicle, and pedestrian, and identifies pedestrian traffic light status. The indication of where and when to cross the road is provided to the visually impaired based on the crosswalk scene understanding. Our proposed system is implemented on a head-mounted mobile device (SensingAI G1) equipped with an Intel RealSense camera and a cellphone, and provides surrounding scene information to visually impaired individuals through audio signal. To validate the performance of the proposed system, we propose a crosswalk scene understanding dataset which contains three sub-datasets: a pedestrian traffic light dataset with 7447 images, a dataset of key objects on the crossroad with 1006 images and a crosswalk dataset with 3336 images. Extensive experiments demonstrated that the proposed system was robust and outperformed the state-of-the-art approaches. The experiment conducted with the visually impaired subjects shows that the system is practical useful.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/tnsre.2021.3096379

Cites

40 of 40 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 The examples of crosswalk dataset. (a) negative samples; (b) positive samples, which are respectively complete, broken, waterlogged and badly-lighted crosswalk from left to right.
Fig. 1 The examples of crosswalk dataset. (a) negative samples; (b) positive samples, which are respectively complete, broken, waterlogged and badly-lighted crosswalk from left to right.CC-BY-NC-ND · From the PDF
Fig. 2 The head-mounted assistive device for the blind (SensingAI G1): an RGB-D camera, a bone conduction headset, and a blind assistant app integrates the proposed system and GPS navigation.
Fig. 2 The head-mounted assistive device for the blind (SensingAI G1): an RGB-D camera, a bone conduction headset, and a blind assistant app integrates the proposed system and GPS navigation.CC-BY-NC-ND · From the PDF
Fig. 3 The crosswalk scene understanding system. The crosswalk, pedestrian, vehicle and PTL are detected. The distance of key objects(pedestrian, vehicle, PTL…) are measured. VTL:vehicle traffic light; PTL:pedestrian traffic light.
Fig. 3 The crosswalk scene understanding system. The crosswalk, pedestrian, vehicle and PTL are detected. The distance of key objects(pedestrian, vehicle, PTL…) are measured. VTL:vehicle traffic light; PTL:pedestrian traffic light.CC-BY-NC-ND · From the PDF
Fig. 4 The framework of crosswalk detector.
Fig. 4 The framework of crosswalk detector.CC-BY-NC-ND · From the PDF
Fig. 5 Diagram of candidate extraction and candidate analysis: (a) input image; (b) HSV image; (c) gray image; (d) crosswalk mask; (e) white region(candidates); (f) result of median filter; (g) result of Canny edge detection; (h) result of Hough transform.
Fig. 5 Diagram of candidate extraction and candidate analysis: (a) input image; (b) HSV image; (c) gray image; (d) crosswalk mask; (e) white region(candidates); (f) result of median filter; (g) result of Canny edge detection; (h) result of Hough transform.CC-BY-NC-ND · From the PDF
Fig. 6 The sketch of θ , which ranges in [0, 10 ◦ ].
Fig. 6 The sketch of θ , which ranges in [0, 10 ◦ ].CC-BY-NC-ND · From the PDF
Fig. 7 Diagram of pedestrian traffic light recognition: (a) RGB image; (b) candidate; (c) gray image; (d) HSV image; (e) traffic light mask; (f) the shape of traffic light.
Fig. 7 Diagram of pedestrian traffic light recognition: (a) RGB image; (b) candidate; (c) gray image; (d) HSV image; (e) traffic light mask; (f) the shape of traffic light.CC-BY-NC-ND · From the PDF
Fig. 8 The AspectRatio of different shapes of traffic light: (a-d) VTL; (e-f) PTL; (g) extinguished traffic light.
Fig. 8 The AspectRatio of different shapes of traffic light: (a-d) VTL; (e-f) PTL; (g) extinguished traffic light.CC-BY-NC-ND · From the PDF
Fig. 9 The qualitative comparison of crosswalk detection algorithms in different scenarios.
Fig. 9 The qualitative comparison of crosswalk detection algorithms in different scenarios.CC-BY-NC-ND · From the PDF
Fig. 10 The detected results by YOLOv4 (a) and our method (b).
Fig. 10 The detected results by YOLOv4 (a) and our method (b).CC-BY-NC-ND · From the PDF
Fig. 11 The laser rangefinder and its use cases. The distances labeled in red are the results measured by our system and the distances marked in dark blue are that measured by the rangefinder which represent the ground truth. (a) The laser rangefinder; (b) The measurement case of PTL; (c) The measurement case of vehicle.
Fig. 11 The laser rangefinder and its use cases. The distances labeled in red are the results measured by our system and the distances marked in dark blue are that measured by the rangefinder which represent the ground truth. (a) The laser rangefinder; (b) The measurement case of PTL; (c) The measurement case of vehicle.CC-BY-NC-ND · From the PDF
Image 12
CC-BY-NC-ND · From the PDF
Fig. 13 Some examples of image output detected by the proposed system.
Fig. 13 Some examples of image output detected by the proposed system.CC-BY-NC-ND · From the PDF
Fig. 14 The testing routes and crossroad scenes.
Fig. 14 The testing routes and crossroad scenes.CC-BY-NC-ND · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 41 references

Research works printed but not linked: 4

  1. huang, 2017, An improved method of zebra crossing detection based on bipolarity, Applied Soft Computing, 12, 202
  2. chen, 2019, Zebra crossing recognition method based on edge frature and Hough transform, J Zhejiang Univ Sci, 6, 476
  3. 2021, World Health Organization
  4. roters, 2011, Pedestrian Lights Database

Cited by

See all 57

Metadata

Authors
Tian, S., Zheng, M., Zou, W., Li, X., & Zhang, L.
Affiliations
  • Shenzhen University (CN)
  • Institut National des Sciences Appliquées de Rennes (FR)
Venue
IEEE Transactions on Neural Systems and Rehabilitation Engineering
Published
2021
Volume
29
Pages
1478-1486
ISSN
  • 1534-4320 (ISSN-L)
  • 1558-0210
OpenAlex
W3180901291
Cited by
48
References
41
Citations by year
2026: 4 · 2025: 12 · 2024: 11 · 2023: 13 · 2022: 8
FWCI
2.88
Type
journal article
Language
en
Open access
Locations
  • IEEE Transactions on Neural Systems and Rehabilitation Engineering (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • HAL (submittedVersion, other-oa)
Funders
  • National Natural Science Foundation of China (61771321, 61871273, 61872429)
  • Key Project of DEGP (2018KCXTD027)
  • Natural Science Foundation of Guangdong Province, China (2020A1515010959)
  • Natural Science Foundation of Shenzhen (JCYJ20200109105832261)
  • Interdisciplinary Innovation Team of Shenzhen University
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Health Sciences › Health Professions › Occupational Therapy › Assistive Technology in Communication and Mobility
  • Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
Keywords
visually impaired people, independent mobility, head-mounted devices, audio feedback, object detection, assistive technology
MeSH terms
Accidents, Traffic, Pedestrians, Humans, Visually Impaired Persons, Persons with Visual Disabilities

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Dynamic Crosswalk Scene Understanding for the Visually Impaired. https://app.cochat.ai/papers/10.1109/tnsre.2021.3096379
The paper
Tian, S., Zheng, M., Zou, W., Li, X., & Zhang, L. (2021). Dynamic Crosswalk Scene Understanding for the Visually Impaired. IEEE Transactions on Neural Systems and Rehabilitation Engineering. https://doi.org/10.1109/tnsre.2021.3096379

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.