302 images
Dataset Image
Dataset Image
Dataset Image
Dataset Image
Dataset Image
Dataset Image
Dataset Image

aircraft skin defects (merged) Computer Vision Project

DDIISc

Updated 2 years ago

165

views

0

downloads
Tags
Object Detection
Classes (5)
crack dent
missing-head
paint-peel-off
scratch
Description

This is a custom dataset created with manually clicked photographs of 3 different types of aircraft. The images are labelled for 5 different classes of surface defects on aircraft skin viz.

  1. Crack
  2. Dent
  3. Scratch
  4. Paint-peel-off
  5. Missing-head (rivets, fastners, screws etc)

The dataset consists of total 372 images with distribution as follows: Hawker Hunter - 276 HT2 - 70 Pushpak - 76

The dataset has been created for implementation as part of M.Tech (AI) thesis of the author at IISc, Bengaluru.

Note: The author couldn't find any detailed / relevant dataset available on the internet. Thus this dataset aims at providing a readymade repo of such images for future use as well. Please feel free to use the images for your own model by citing the requisite credits to the author.
Supervision

Build Computer Vision Applications Faster with Supervision

Visualize and process your model results with our reusable computer vision tools.

Cite This Project

LICENSE
CC BY 4.0

If you use this dataset in a research paper, please cite it using the following BibTeX:

                        @misc{
                            aircraft-skin-defects-merged_dataset,
                            title = { aircraft skin defects (merged) Dataset },
                            type = { Open Source Dataset },
                            author = { DDIISc },
                            howpublished = { \url{ https://universe.roboflow.com/ddiisc/aircraft-skin-defects-merged } },
                            url = { https://universe.roboflow.com/ddiisc/aircraft-skin-defects-merged },
                            journal = { Roboflow Universe },
                            publisher = { Roboflow },
                            year = { 2023 },
                            month = { apr },
                            note = { visited on 2024-11-21 },
                            }

Similar Projects

See More