#
Page tree
Skip to end of metadata
Go to start of metadata

Without any knowledge of the information nested in a HS cube, targeted investigations are typically misleading (like spectra selection).
An unsupervised method to describe dominant spectroscopic information expressed as colour information can help to understand the general nature of existent information. Therefore, by knowledge of the preview information, targeted investigations are possible.


Function

The Preview method gives an unsupervised approach to extract information from HS cubes. The Preview method can be seen as a black box: The input is a hyperspectral cube and the output image a color image. Spectroscopic differences are shown in different color, while spectroscopic similarities are shown in similar colors. As big as the color differences, are the spectroscopic differences in the spectra. Therefore 2 objects colored in very distinct colors are spectroscopically distinct and can be expected to be chemically distinct. By also investigating preprocessing often an appropriate preprocessing method can be estimated. The preprocessing method might be fit for modelling if the resulting Preview image reflects the expectations of the user.

User Interface

Workflow

  • Select a cube and study its preview visualization
  • Apply pre-processing and in parallel compare results with your expectations
  • Crop the scene to focus only on objects of interest.
  • Set the scaling to selected spectra to arrange the color contrast in the result
  • Save the model for later usage (e.g. to set up the live streaming)

Hints

Since the result is influenced by the value dynamic in the data, try to avoid discontinuities like reflections etc.
Use the Preview method with different configurations for the hyperspectral preprocessing. The preprocessing method best depicting the expected behavior might by the best choice for further processing.

Mentioned in:

Found 7 search result(s) for "preview method".

Page: Explore Hyperspectral Information (Manuals)
... extract information from HS cubes. See further information here: CCI Preview Method. Statistical Minimum It results in the statistical minimum of the spectral response per object pixel ... ...
May 26, 2020
Page: Crop Hyperspectral Data (Manuals)
... starting the Crop process ensure the selected hyperspectral cube is shown prominently. The CCI Preview method is used for visualization of the hyperspectral cube. Try different preprocessing options to get a good ... ...
May 28, 2020
Page: Merge Hyperspectral Data (Manuals)
... starting the Merge process ensure the selected hyperspectral cube is shown prominently. The CCI Preview https://wiki.perceptionpark.com/display/MAN/CCIPreviewMethod method is used for visualization of the hyperspectral cube. Try different preprocessing options to get a good ... ...
Jun 02, 2020
Page: Exploratory analysis of plastics (Manuals)
... information behind each pixel in the scene. plasticspreviewIntens.png The above image shows the output of the preview method gained from reflectance data, while the next image shows the output of the preview ... ...
Jun 16, 2020
Page: Select Spectra (Manuals)
... Spectra process ensure the selected hyperspectral cube is shown most prominently. The CCI Preview method is used for visualization of the hyperspectral cube. Try different preprocessing options to get a good ... ...
May 28, 2020
Page: Model Hyperspectral Information (Manuals)
... from the Training Data Panel. In the Training Data Panel you should see the Preview of the selected hyperspectral data Please refer to the Preview Method documentation for further information. Consider cropping the data to get a reasonable data size ... ...
May 27, 2020
Page: Hyperspectral imaging step by step (Manuals)
... Perception Studio's Model perspective → Select the data for the modelling process → Choose a modelling method available in the ribbon → Develop a model and save it for later usage Depending on the application different ...
Jul 07, 2020



© 2019 by Perception Park GmbH
The content of this page and any attached files are confidential and intended solely for the addressee(s). Any publication, transmission or other use of the information by a person or entity other than the intended addressee is prohibited. If you receive this in error please contact Perception Park and delete copied material. Perception Park GmbH, Wartingergasse 42, A-8010 Graz; FN 400381x

  • No labels