py-XVis

Python implementation for XVis Toolbox release with the book Computer Vision for X-Ray Testing. Originally implemented in Matlab by Domingo Mery for the first edition of the book. This package is part of the second edition of the book Computer Vision for X-Ray Testing (November 2020).

Requirements

• Python 3.6 or higher
• numpy < 1.19
• matplotlib >= 3.3.2
• scipy >= 1.5.2
• pyqt5 >= 5.15.1
• pybalu >= 0.2.9
• opencv-python = 3.4.2
• opencv-contrib-python = 3.4.2
• tensorflow >= 2.3.1
• scikit-learn >= 0.23.2
• scikit-image >= 0.17.2
• pandas >= 1.1.2

Instalation

The package is part of the Python Index (PyPi). Installation is available by pip:

pip install pyxvis

Interactive Examples

All examples in the Book have been implemented in Jupyter Notebooks tha run on Google Colab.

Chapter 01: X-ray Testing

• Example 1.1: Displaying X-ray images
• Example 1.2: Dual Energy
• Example 1.3: Help of PyXvis functions

Chapter 02: Images for X-ray Testing

• Example 2.1: Displaying an X-ray image of GDXray

Chapter 03: Geometry in X-ray Testing

• Example 3.1: Euclidean 2D transformation
• Example 3.2: Euclidean 3D transformation
• Example 3.3: Perspective projection
• Example 3.4: Cubic model for distortion correction
• Example 3.5: Hyperbolic model for imaging projection
• Example 3.6: Geometric calibration
• Example 3.7: Epipolar geometry
• Example 3.8: Trifocal geometry
• Example 3.9: 3D reconstruction

Chapter 04: X-ray Image Processing

• Example 4.1: Aritmetic average of images
• Example 4.2: Contrast enhancement
• Example 4.3: Shading correction
• Example 4.4: Detection of defects using median filtering
• Example 4.5: Edge detection using gradient operation
• Example 4.6: Edge detection with LoG
• Example 4.7: Segmentation of bimodal images
• Example 4.8: Welding inspection using adaptive thresholding
• Example 4.9: Region growing
• Example 4.10: Defects detection using LoG approach
• Example 4.11: Segmentation using MSER
• Example 4.12: Image restoration

Chapter 05: X-ray Image Representation

• Example 5.1: Geometric features
• Example 5.2: Elliptical features
• Example 5.3: Invariant moments
• Example 5.4: Intenisty features
• Example 5.5: Defect detection usin contrast features
• Example 5.6: Crossing line profiles (CLP)
• Example 5.7: SIFT
• Example 5.8: feature se;ection
• Example 5.9: Example using intenisty features
• Example 5.10: Example using geometric features

Chapter 06: Classification in X-ray Testing

• Example 6.1: Basic classification example
• Example 6.2: Minimal distance (dmin)
• Example 6.3: Bayes
• Example 6.4: Mahalanobis, LDA and QDA
• Example 6.5: KNN
• Example 6.6: Neural networks
• Example 6.7: Support Vector Machines (SVM)
• Example 6.8: Training and testing many classifiers
• Example 6.9: Hold-out
• Example 6.10: Cross-validation
• Example 6.11: Confusion matrix
• Example 6.12: ROC and Precision-Recall curves
• Example 6.13: Example with intensity features
• Example 6.14: Example with geometric features

Chapter 07: Deep Learing in X-ray Testing

• Example 7.1: Basic neural networks (from skratch)

• Example 7.2: Neural network using sklearn

• Example 7.3: Convolutional Neural Network

• Example 7.4: Pre-trained models

• Example 7.5: Fine tunning

• Example 7.6: Generative Adversarial Networks (GANs)

• Example 7.7: Object detection using YOLOv3

• Example 7.8: Object detection using YOLOv4

• Example 7.9: Object detection using YOLOv5

• Example 7.10: Object detection using EfficientDet

• Example 7.11: Object detection using RetinaNet

• Example 7.12: Object detection using DETR

• Example 7.13: Object detection using SSD

Chapter 08: Simulation in X-ray Testing

• Example 8.1: Basic simulation using voxels
• Example 8.2: Simulation of defects using mask
• Example 8.3: Simulation of ellipsoidal defects
• Example 8.4: Superimposition of threat objects

Chapter 09: Applications in X-ray Testing

• Example 9.1: Defect detection in castings

• Example 9.2: Defect detection in welds