Van de Loosdrecht Machine Vision BV
Company was founded in 2001Core business Machine Vision (Image Processing)
- Development of the image processing system VisionLab
- Image processing algorithms
- Deep learning
- Hyper-spectral imaging
- Genetic algorithms
- Pattern matching
- Algorithms written in ANSI C++
- OpenMP
- OpenCL
- Portable software, running under Windows, Linux and Android on x86, x64, ARM and PowerPC
News
VisionLab interfaces with all major Deep Learning frameworks that have a Python interface. See also
Deep Learning. Lecture
Some deep thoughts on Deep Learning about the transition from Software 1.0 to Software 2.0
Do you think your object detection model doesn't discriminate small overlapping objects adequately? Then have a look at the new publication on CentroidNetV2: A hybrid deep neural network for small-object segmentation and counting, Neurocomputing 423 (2021) p 490-505
New Hyper-Spectral Imaging operators are under development Hyper-Spectral Imaging
Support for SPECIM FX10&FX17, Ximea and Silios hyper-spectral cameras
Support for LUCID polarization cameras
Lecture Introduction to Hyper Spectral Imaging NVPHBV (Dutch society of Pattern Recognition and Image Processing) meeting Spring 2019 at NHL Stenden University
Inauguration speech Jaap van de Loosdrecht, Professor Computer Vision, NHL-Stenden University of Applied Sciences
Download inaguration speech (Dutch)
Thesis Accelerating sequential machine vision algorithms using commodity parallel hardware available for download.
New version of VisionLab V3.47 (January 2020) available in download section:
Do you think your object detection model doesn't discriminate small overlapping objects adequately? Then have a look at the new publication on CentroidNetV2: A hybrid deep neural network for small-object segmentation and counting, Neurocomputing 423 (2021) p 490-505
New Hyper-Spectral Imaging operators are under development Hyper-Spectral Imaging
Support for SPECIM FX10&FX17, Ximea and Silios hyper-spectral cameras
Support for LUCID polarization cameras
Lecture Introduction to Hyper Spectral Imaging NVPHBV (Dutch society of Pattern Recognition and Image Processing) meeting Spring 2019 at NHL Stenden University
Inauguration speech Jaap van de Loosdrecht, Professor Computer Vision, NHL-Stenden University of Applied Sciences
Download inaguration speech (Dutch)
Thesis Accelerating sequential machine vision algorithms using commodity parallel hardware available for download.
New version of VisionLab V3.47 (January 2020) available in download section:
- Interfaces with all major Deep Learning frameworks that have a Python interface, see also Deep Learning
- Multi Core Processing with automatic operator parallelization for more than 170 operators, available for all supported processors
- Scripts and programs will run faster without modification
- Windows 10 x64 version
- Linux x64 and ARM version
- Raspberry PI and ODRIOD C1/XU4 versions
- Android version
- HTML help, see help file
- XIMEA xiAPI camera interface
- USB Video Class (UVC) camera interface
- Basler Pylon camera interface
- IDS uEye camera interface
- SPECIM FX10 & FX17 camera interface (beta)
- LUCID camera interface (beta)
- Tethering with Nikon and Canon DSLR cameras
- Toolbox for using OpenCL kernels from scripts and C++
- SWIG interface, example of using VisionLab directly from C#
- SDK for Visual Studio and GNU C++ compiler
- See change history for all changes
Thesis
Accelerating sequential machine vision algorithms using commodity parallel hardware
- Research project in collaboration with the Limerick Institute of Technology in Ireland and the NHL University in the Netherlands
- This project focus on both multi-core CPUs and GPUs
- Tools used are C++, OpenMP and OpenCL
- Download thesis
- Course material about acceleration vision algorithms using OpenMP and OpenCL