The new version of ScanEx Image Processor v.3.0 has been developed by ScanEx specialists for more than two years and was supplemented with innovation technology solutions for processing and anaysis of space images.
The highlights of the new ScanEx Image Processor v. 3.0 version in basic configuration are as follows:
* Dramatic extension of functional for geometric correction of different data types. Now it is possible to work with strict model for data delivered from SPOT-2/4, RESOURCESAT-1 (IRS-P6), CARTOSAT-1, EROS-А/B, FORMOSAT-2; to verify models of Resourcesat-1 (IRS-P6), Cartosat-1 (IRS-P5) and IKONOS data of GeoOrthoKit format, QuickBird data of QuickBird format and PRISM camera data from ALOS satellite in NITF format using control points based on RPC-coefficients selecting transformation model; to specify elevation matrix using different methods when doing ortho-correction; to save and open GCP’s in diverse projections and coordinate systems.
* A new global 3D-model of the Earth was designed.It is now possible using 3D-model of the Earth to do toponymic search on the Earth surface based on vector layers, to overlay local raster 3D-models, to visualize satellite pass trajectories and their tracks, to record video-clips, etc.
* The image mosaic functional has been changed, enabling to create mosaics coverage for large areas. Universal mosaicking tool provides interactive or automatic tonal adjustment of space images with the option to do non-linear color balancing.
* The user-defined data library has been added. The data library is the catalog of raster data records, enabling to select from the list the image for the specific territory, to assess the overplay percentage and to load into program forming a unified raster coverage.
* A new functional module has been added for classification methods. The new version of ScanEx Image Processor enables to select the classification method depending on the assigned task and source data, as well as offers an additional toolkit for statistic processing of obtained results. If there are not templates the software suggests that the known algorithm of ISODATA unsupervised classification is used. The specific of using this algorithm in ScanEx Image Processor v. 3.0 is in the possibility to include a vector layer with classification boundaries for training. Classification with a “teacher” is done thru neural network of forward propagation using fuzzy logic when specifying a probability threshold of mixed pixels distribution.
* The functional module for raster data visualization has been improved making it possible to to set up the transparency effect for any raster layer; to display raster layer within a vector mask; to fine-tune display parameters using additional contrast and brightness adjustment tools and to save settings in external files.
* New functional possibilities have been introduced in vector editor.
* Operations with attributive table of vector layer are now available, which is used to modify records, to search, select and sort out vector objects, etc.
In addition to basic configuration most popular software applications experienced essential modifications:
3D-modeling and visualization module
This module new features are: usage of vector layers and of exported into vector layer GPS receiver data as camera flyby trajectory or 3D-object; quick creation of typical 3D-objects using contour line of vector maps and textures with their interactive editing; streamlined application of simple vector primitives (polylines, points and polygons) on a 3D-model; possibility of operations with attributive data, stored in the database of 3D-vector objects; more user-friendly and extended toolkits for navigation and control.
Thematic Pro
It includes updated object-thematic classification and thematic interpretation of multi-spectral optical and radar data, as well as classification of data based on extended neural network algorithm of thematic classification and interpretation and its development into GTM algorithm.
In object-orientated classification and multi-spectral imagery interpretation it became possible to take into account the object’s geometry during segmentation; classification of segmentation results is done using discrimination analysis, plus interactive thematic supervised classification for segmentation results. Neural network classification algorithm and GTM enable to use pixel and object data for images classification, to build intuitive structure of regions’ classes, to teach the neural network within the limits of model samplings. Different algorithms of post-processing and work with hierarchy structure of regions are also built into the classification.
The new version of ScanEx Image Processor v.3.0 has been successfully tested and used as an irreplaceable basis for processing and analysis of spatial data in several major projects of ScanEx company.
Starting in 2008, four blocks of training courses were taken on ScanEx Image Processor v.3.0 software. In particular, the course for forest management enterprises - part of Federal State Unitary Enterprise “Roslesinforg” – that introduced ScanEx’s software into practice. Within the frames of this partnership over 40 people were trained and over 50 software product licenses granted, both local and network.
In many organizations the new version of ScanEx Image Processor v.3.0 software has been installed and successfully used.
More information about the software features is available by downloading the free demo-version and its leaflet from the ScanEx’s web-site at the following link
http://www.scanex.ru/en/software.
