Hyperspectral Image Processing Software for THz Imaging Data

Prozessschema 3w


Advanced Features and Benefits:

  • One-of-a-kind image processing tool
  • Based on statistical methods
  • Analysis of THz and other spectroscopic image data
  • Flexible data import
  • Pre-processing, analysis, visualization, and highlighting of information

Basic Edition Extended Edition Database Edition (coming soon)
Detection of suspicious pixels 4D data model User-definable spectral databases
Image stack 64-bit architecture Support of third-party databases
Basic set of mathematical methods Multi-sensor support Library search
Vertex component analysis ImageLab script automation  
Principal component analysis KNN classifiers  
Cluster analysis Random forest classifiers  
PLS discriminant analysis Multiprocessing support  
Similarity maps    
Baseline correction    
Spectral descriptors    
3D image display    
Maximum noise fraction transform    


ImageLab is a one-of-a-kind image processing tool for the analysis of data from various spectroscopic imaging techniques such as THz, optical, UV/VIS, IR, Raman, or mass spectrometry. Based on multivariate statistical methods it enables the user to analyze and classify acquired hyperspectral images, to merge them with maps of physical properties and to overlay them with conventional high-resolution color photos. A chemometrics toolbox, visualization of the image data and highlighting of analytical information round up the software package to an indispensable instrument for in-depth analysis of spectroscopic image data.

ImageLab is now part of Menlo Systems’ imaging extension TERA Image. Ask us for possibilities to upgrade your THz imaging system with our unique software tool!


FREE trial license available here.


  • TERA Image Extension Unit

    Product Code: TERA Image


    For THz imaging applications our fully automated extension TERA Image can be integrated.


  • ImageLab: Extended Edition THz Image Analysis

    Product Code: ImageLab


    Hyperspectral Image Processing Software for THz Imaging Data


Antenna and chip structure in a key card

Photo and transmitted THz signal distribution, selected for the spectral range of 1.5 – 2.5 THz:

The microchip, antenna windings, and the wiring are visible due to the increased image resolution of the high-frequency range.



K-means cluster analysis ClassMap (k = 8):

The k-means algorithm breaks down the input data set into k partitions (culsters). The picture below shows an automatic analysis of the 0.7-2.1 THz region. Due to different diffraction of linearly polarized THz radiation the vertically and horizontally oriented structures of the antenna induce different spectral properties. The different material components of the key card such as plastic, semiconductor chip, or metal are clearly recognized.




Identification of biomolecules

Photo of a sample consisting of three different biomolecules, α-lactose, L-tyrosine, L-glutamine, and transmitted THz signal distribution, filtered for the range of 1.5 - 2 THz:

Despite comparable thickness of each layer, the different sample regions exhibit different transmission intensity.



K-means cluster analysis (k = 4):

The algorithm uses the significant spectral components of the three substances for their identification and maps their occurrence within the mixture.

MENLO_SYSTEMS_Terahertz_Cluster map_Biomolekuele

Menlo Expert Milan Oeri 2022
Dr. Milan Öri
Your direct line to our expert

Data sheets

Product literature

Ordering information

  • Product Code
  • ImageLab

+49 89 189166 0

Menlo Systems, Inc.
+1 973 300 4490

Feedback? Click here