Published September 9, 2025 | Version 1.0.0
Dataset Open

Cone-Beam Computed Laminography Dataset of a Keyboard Circuit Board

Description

Summary

This dataset contains X-ray projection images of a keyboard circuit board acquired using a cone-beam computed laminography (CBCL) scanner. The dataset also includes the corresponding .pca metadata file, which contains the essential scanning parameters required for image reconstruction.

 

Description

Sample Information:

The sample is a section of a keyboard circuit board that was affixed to a flat carbon panel using transparent plastic tape to ensure stability during the scanning process. Maximal dimensions of the circuit board are 39 x 25 x 3,5 mm.

Scanner:

The computed laminography was performed using a Phoenix Microme|x Neo scanner from Waygate Technologies. The  inspection was managed using the Phoenix|x-ray x|act software, with scanning performed using its PlanarCT module.  This scanner is equipped with flat panel detectors and utilizes a cone-beam geometry for the X-ray source. The scanner is capable of scanning at 45- and 60-degree tilt angles and is located at the LUT School of Energy Systems (LES) at the Lappeenranta campus of Lappeenranta-Lahti University of Technology LUT.

For more technical specifications and details on the scanner, please refer to the Waygate Technologies product page: https://www.bakerhughes.com/waygate-technologies/industrial-radiography-and-ct/electronics-xray-inspection-axi/micromex-nanomex-neo 

Scan Settings:

The scan was conducted at a 45-degree tilt angle. The object was rotated 360 degrees horizontally on the carbon panel to acquire a full set of projections. The X-ray source was operated at a voltage of 100 kV and a current of 80 µA, resulting in a total power of 8 W. The detector exposure time was set to 200 ms, and a total of 758 projection images were acquired. To ensure accurate data acquisition, several calibrations were performed before the scan. First, the center of rotation was precisely determined. Following this, the manipulator was calibrated, which involved establishing the exact position of the manipulator's axis relative to the detector. Finally, the detector was calibrated by performing flat-field and dark-field corrections. This process determines the offset image, the gain image, and the mean free ray gray scale value. More information about the scan settings can be found from the .pca metadata file. 

Abbreviations in the metafile:

  • FOD:  Focus-to-Object Distance in mm (same as Source-to-Origin Distance)
  • FDD: Focus-to-Detector Distance in mm (same as Source-to-Detector Distance)
  • Tilt: Angle of the X-ray source relative to the object's axis of rotation in degrees
  • NumberImages: Total  number of projection images (angles)
  • DimX: Horizontal pixels of the projection
  • DimY: Vertical pixels of the projection
  • PixelsizeX: Detector pixel width in mm
  • PixelsizeY: Detector pixel height in mm
  • TimingVal: Exposure time of the detector in seconds
  • Voltage: Voltage of the source in kV
  • Current: Current of the source in µA
  • FreeRay: Initial X-ray photon intensity determined by the scanner software

 

Research Group

The dataset was produced by the Applied Mathematics Research Unit in the Department of Computational Engineering at LUT University, Finland: https://www.lut.fi/en/research-groups/applied-mathematics-0

Additional Links

For processing this dataset and performing reconstructions, we recommend using the Core Imaging Library (CIL), an open-source framework for tomographic image reconstruction. A practical example of working with CBCL data using CIL can be found in this Jupyter notebook:  https://github.com/TomographicImaging/CIL-Demos/blob/main/demos/2_Iterative/05_Laminography_with_TV.ipynb 

Contact Details

For more information about the dataset, please contact Tommi Heikkilä at tommi.heikkila[at]lut.fi or Toni Karvonen at toni.karvonen[at]lut.fi.

Files

Circuit-Board-Image.jpeg

Files (910.5 MB)

Name Size Download all
md5:560bd22d7737810a1c51c9e137a9d60b
909.8 MB Preview Download
md5:108ee9e0ecf58d77dad04b66e6d3f6f1
784.1 kB Preview Download

Additional details

Funding

Research Council of Finland
Flagship of Advanced Mathematics for Sensing, Imaging and Modelling 359183
Research Council of Finland
Inference and approximation under misspecification 368086