Object Analyst

You use Object Analyst, an object-based image-analysis (OBIA) module, to segment an image into objects for classification and analysis. By segmenting an image into objects, analysis can be both simplified and made more sophisticated. That is, by using Object Analyst to perform object-based classification, human-visual interpretation of images can be augmented by using the software to do much of the preliminary work in creating and determining which shapes, of given sizes, textures, and so forth, are of interest.

Object Analyst is for use primarily with very-high-resolution (VHR) imagery; however, you can use Object Analyst with any imagery that meets the necessary criteria. That is, you can use imagery of lower resolution, of various resolutions, and that is in an input format supported by Focus.

Note: While you can use various formats as input, the output file produced by Object Analyst is exclusively in PCIDSK format (.pix).

Features and benefits

The major advantages of Object Analyst are as follows:

Process-driven, flexible workflow
The all-in-one interface guides you through segmenting your imagery, calculating attributes, creating training sites, classification (including creating custom rules for classification), reforming shapes, working with classes, and performing an accuracy assessment.
Easily delineate homogenous regions
A proprietary image-segmentation algorithm delineates homogenous regions based on statistical region-growing.
Fast-and-easy attribute calculation
Quickly and easily calculate statistical and geometrical attributes for segments (polygons). The process is not limited to the input raster (scale and cell size), but can also be performed on rasters of various resolutions and data ranges.
Advanced analysis technology
Object Analyst uses machine-learning technology Support Vector Machine (SVM) and traditional maximum likelihood classifier (MLC) for supervised classification of vector objects. The k-means algorithm is also used to split data into natural clusters to aid data exploration and understanding.
Use classification rules you define
Create custom rules for classification to suit your data and the analysis you want to perform based on the calculated attributes.

For more information, see: