This page uses cookies. For details and possible settings refer to our Privacy Policy.
Continuing to use this page means accepting the processing of cookie files.

Accept (hide info)
POL ENG GER 中文(繁體) 中文(简体)

Machine Vision Library for C++ and .NET


Adaptive Vision Library is a machine vision library for C++ and .NET programmers. It provides a comprehensive set of functions for creating industrial image analysis applications – from standard-based image acquisition interfaces, through low-level image processing routines, to ready-made tools such as template matching, measurements or barcode readers.

Code Samples Documentation

The functions available in Adaptive Vision Library closely correspond to the filters of Adaptive Vision Studio. Therefore, it is possible to prototype your algorithms quickly in a graphical environment and then translate them to C++ or .NET, or even generate the C++ code automatically.

Adaptive Vision Library gives you instant access to the highest quality, well optimized and field-tested code that you need for your machine vision projects!



In Adaptive Vision Library careful design of algorithms goes hand in hand with extensive hardware optimizations, resulting in performance that puts the library among the fastest in the world. Our implementations make use of SSE/AVX/NEON instructions and parallel computations on multicore processors.

Modern Design

All types of data feature automatic memory management, errors are handled explicitly with exceptions and optional types are used for type-safe special values. All functions are thread-safe and use data parallelism internally, when possible.

Simplicity & Consistency

The library is a simple collection of types and functions, provided as a single DLL/SO file with appropriate headers. For maximum readability, functions follow consistent naming convention (e.g. SmoothImage, RotateVector). All results are returned via reference output parameters so that many outputs are always possible.


Free for everyone
fundamental image processing and 2D geometry
For companies developing their own algorithms but who need a strong base library (min. order: 25)
For companies providing solutions, looking for ready-made tools for industrial applications
Runtime price for integrators (standard)
suitable for single-camera vision systems; a development license is required; contact sales for details
0€ 300€ 800€$
Runtime price for integrators (multi-threaded)
suitable for multi-camera vision systems, 8-core CPU; a development license is required; contact sales for details
0€ 525€ 1400€
Unified third party image acquisition
Allied Vision, Basler, Dahua, FLIR, Hikvision, IDS, Lucid Vision, MATRIX VISION, NET, ThorLabs, XIMEA, …
Yes Yes Yes
Fundamental image processing
image smoothing, image morphology, spatial transforms, pixel-wise operators, gray-scale thresholding, file I/O
Yes Yes Yes
Comprehensive library for 2D geometry Yes Yes Yes
Region of interest functions Yes Yes Yes
Region analysis
Highly optimized with RLE representation
No Yes Yes
Advanced geometrical algorithms
caliper diameter, fitting primitives to points etc.
No Yes Yes
Histogram analysis No Yes Yes
Basic computer vision
advanced thresholding, all color space conversions, LUT, Hough transform, watersheds, distance transform, image sharpness, extrema analysis, corner detection
No Yes Yes
Advanced computer vision
image inpainting, high-quality 2D edge detection, morphology with arbitrary kernels, spatial transforms based on paths, color distance analysis, golden template, maximally stable feature regions, point pattern matching point, advanced contour (path) analysis, geometrical fitting with outliner suppression, 1D profile analysis, data classification tools (MLP, SVN, KNN, PCA), advanced statistical analysis
No Option Yes
GenICam + GigE Vision
Our GigE Vision consumer library is certified by AIA
No Option Yes
Camera calibration
Remove lens distortion, calculate real-world coordinates
No Option Yes
Barcode reading
EAN8, EAN13, UPCA, UPCE, Code39, Code93, Code12, Interleaved2of5, Pharmacode
No Option Yes
Datacode reading
Data Matrix, QR Code
No Option Yes
OCR No Option Yes
Template matching
NCC or edge-based
No Option Yes
Metrology tools
1D edge detection + shape fitting
No Option Yes
Photometric stereo No Option Yes
3D machine vision No No Yes
Special tools
e.g. fast image segmentation, fast line segment detection
No No Yes
Deep Learning Add-on
feature detection, anomaly detection, object classification, instance segmentation and point location
No No Option
C++ interface
available as a DLL or SO + headers
Yes Yes Yes
C# interface
available as a .NET assembly
No Yes Yes
.NET controls for high speed image display
based on DirectX, Windows-only
No Yes Yes
.NET controls for editing shapes and models
WinForms, Windows-only
No Yes Yes


How does AVL compare to OpenCV?

OpenCV is a great open source library maintained by a non-profit foundation. In terms of fundamental image processing, OpenCV provides similar functionality to our AVL Lite. Major differences arise when it comes to higher level algorithms. OpenCV focuses on such topics as face recognition, augmented reality or motion segmentation. On the other hand, AVL provides ready-made tools for shape analysis, template matching or code reading which are more specific to industrial machine vision applications. Some unique features of AVL also are: blob analysis, any-shape regions of interest and support for industrial image acquisition interfaces. For more details see: Libraries Comparison.

I do not need entire library but some of the tools would be useful.

With minimum order of 25 runtime licenses we can offer you an OEM contract. You can select one or two modules that you are interested in and we will provide you that with a discount. Please contact our team for details.

Are there any discounts for volume?

Some of our customers buy hundreds or even thousands of runtime licenses per year. These we will provide with discounts. Please contact our sales for details.

How does your the licensing work?

We offer development and runtime licenses. You need at least one development license and then one runtime for each system you deploy. The price of the runtime version depends on your quantity and hardware setup. Licenses for standard PC computers or embedded systems are cheaper than licenses for computers that can run many inspections in parallel. Our business model here assumes that we give you a product that greatly accelerates development of your vision systems, and the more successful you are the more we are rewarded too.


Adaptive Vision Library C++ can be used with Microsoft Visual Studio C++ (2013-2019) and GCC compilers. It runs on industrial PCs as well as on embedded devices such as:

  • Raspberry Pi 3 - Raspbian
  • Wandboard Quad - Ubuntu
  • NXT Vegas
  • Tegra TX2 - Jetpack
  • UltraZed-EG - Petalinux