Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2026
|
| Online Access: | https://doi.org/10.5281/zenodo.19286465 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p>This dataset contains 10000 road scene images collected from forward-facing driving videos on national highways in India. Each image presents a combined view consisting of the original frame and the corresponding processed output generated using a lane center deviation (LCD) algorithm. The left side of each image represents the original road scene, while the right side shows the detected lane boundaries, estimated lane center, and the computed deviation of the vehicle from the lane center. The dataset is generated using a classical image processing pipeline implemented in Python with OpenCV, including grayscale conversion, Gaussian filtering, edge detection, region of interest selection, and Hough transform-based lane detection. All images are stored in a single directory with a consistent naming convention and resolution. The dataset enables direct visual comparison between input images and their corresponding lane detection outputs within the same frame. This dataset can be used for training, validation, and benchmarking of computer vision and machine learning models for lane detection, lane positioning, and Advanced Driver Assistance Systems (ADAS).</p>