Automatic defect recognition in X-ray testing using computer vision
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
xnet
.DS_Store
MedianDetection.m
README.md
casting.png
exp_test.m
exp_train.m
imdb.mat
imdb_readme.txt
opts.mat
wacv_classifier.m
wacv_cldef.m
wacv_demo.m
wacv_examples.m
wacv_fx.m
wacv_fxdef.m
wacv_sliwin.m
wacv_sparse.m
wacv_test.m

README.md

Xdefects

Automatic defect recognition in X-ray testing using computer vision

Requirements

This code needs the following toolboxes:

Original images are from GDXray

http://dmery.ing.puc.cl/index.php/material/gdxray/

  1. Dataset:

    see imdb.mat and imdb_readme.txt

  2. Code for classical feature extraction and classification:

    see wacv_demo.m and wacv_examples.m in this example the following pretrained nets must be included in folder nets: imagenet-caffe-alex.mat imagenet-vgg-f.mat imagenet-vgg-verydeep-16.mat imagenet-googlenet-dag.mat imagenet-vgg-m-2048.mat imagenet-vgg-verydeep-19.mat please see vlfeat.org for downloading these files.

  3. Code fox Xnet:

    see folder xnet and use code xnet_main.m for training and testing.

  4. Sliding windows for detection in one X-ray image:

    see wacv_sliwin.m

Reference

Mery, D.; Arteta, C.: Automatic Defect Recognition in X-ray Testing using Computer Vision. In 2017 IEEE Winter Conference on Applications of Computer Vision, WACV2017.

(c) 2017 - Domingo Mery and Carlos Artera