GeoRaster can be natively stored in JPEG 2000 Compression now (see JPEG 2000 Compression of GeoRaster Objects). GDAL can be used to load and export JPEG 2000 files with or without decompression (see Loading JPEG and JPEG 2000 Images Without Decompression).

GeoRaster includes two new procedures for JPEG 2000 compression and decompression: SDO_GEOR.compressJP2and SDO_GEOR.decompressJP2. Most GeoRaster functions support reading or writing of JP2 compressed GeoRaster objects, including key operations such as SDO_GEOR.getRasterSubset, SDO_GEOR.subset, SDO_GEOR.rectify, large scale mosaicking, virtual mosaic, and raster algebra.

Mosaicking supports new color balancing methods: statistic matching and histogram matching either using a reference image or automatically detecting and using overlapping area (see Color Balancing During Mosaicking and the STATISTICMATCHING and HISTOGRAMMATCHING values for colorBalance in the mosaicParam Keywords table for SDO_GEOR_AGGR.mosaicSubset). The previous limit of eight overlapping GeoRaster objects for a pixel in mosaicking operation is removed.

GeoRaster includes a new image processing package, SDO_GEOR_IP, which supports linear stretching, piecewise stretching, equalization, normalization, histogram matching, dodging, and filtering. See SDO_GEOR_IP Package Reference for details about the subprograms in that package.

GeoRaster also supports warping (see Image Warping and SDO_GEOR.warp) and affine transformation (see SDO_GEOR.affineTransform).

All new functions are parallelized.

GeoRaster adds over 25 new operators to the Raster Algebra language, including conditionalExpr (IF-THEN-ELSE), ^ (XOR), % (MODULO), POWER, FACTORIAL, and local statistical operators.

GeoRaster provides new raster algebra functions: SDO_GEOR_RA.stack to perform stack statistical analysis, and SDO_GEOR_RA.diff and SDO_GEOR_RA.over to perform logical operations.

The previous limit of eight GeoRaster objects in one raster algebra operation is removed.

Several new core functions and enhancements are provided:

• GeoRaster provides a set of new memory control functions to allow users to control how much memory to use during GeoRaster operations: SDO_GEOR_UTL.getMaxMemSize, SDO_GEOR_UTL.setMaxMemSize, SDO_GEOR_UTL.getReadBlockMemSize, SDO_GEOR_UTL.setReadBlockMemSize, SDO_GEOR_UTL.getWriteBlockMemSize, and SDO_GEOR_UTL.setWriteBlockMemSize. Memory can be set to larger values to improve processing performance (see Managing Memory to Improve Performance).

• The new SDO_GEOR.getCellValues, SDO_GEOR.evaluateDoubles, SDO_GEOR.changeCellValues, and SDO_GEOR.getBitmapMaskValues functions allow multiple points to be queried or updated in a single function call, thus improving performance.

• GeoRaster provides two additional statistic analysis functions: SDO_GEOR.generateAreaWeightedMean to compute area weighted statistical mean value for the cells and sub-cells within a specific window, and SDO_GEOR_UTL.calcSurfaceArea to calculate the three–dimensional (3D) surface area represented by digital elevation model (DEM).

• Two color ramping functions, SDO_GEOR_UTL.generateColorRamp and SDO_GEOR_UTL.generateGrayRamp, allow you automatically generate a color gradient or grayscale gradient for grids and images.

• The new SDO_GEOR.generateBitmapPyramid procedure automates pyramiding on bitmap GeoRaster objects, which significantly improves the pyramid quality. (See Bitmap Pyramiding.)

• SDO_GEOR.getRasterSubset is enhanced to return cell values in an array directly.

• The performance of repeated single-cell query (calling SDO_GEOR.getCellValue repeatedly) over the same GeoRaster object is significantly improved.

• Rectification, JPEG and DEFLATE compression, and statistics generation operations are parallelized.

• For output GeoRaster objects, the default block size in storageParam is changed from 256 x 256 to 512 x 512. (The available parameters and keywords are described in Storage Parameters.)

All Release 12.2 server-side new functions and procedures are supported in the GeoRaster Java API. See Oracle Spatial and Graph Java API Reference for details.

GeoRaster also provides a new Virtual Mosaic Java API to enable faster integration and support advanced visualization applications. See Oracle Spatial and Graph Java API Reference for details.

The GeoRaster ETL and Viewer are integrated into one tool for ease of deployment and use. The virtual mosaic display is significantly enhanced in the Viewer. The GeoRaster Tool provides a new GUI to automate some PL/SQL tasks, such as creating GeoRaster tables, generating pyramids, and rectifying images for demonstration purposes.

Also, GDAL is distributed with Oracle Spatial and Graph. See Using GDAL from the Spatial and Graph Installation.

In addition to WMS, GeoRaster supports the Open Geospatial Consortium (OGC) Web Coverage Services (WCS). You can use WCS to publish GeoRaster objects in the database and allow users to retrieve the raster data over the web, including subsetting, reprojection, and support for many file formats. ( See GeoRaster Spatial Web Services.)

The SDO_GEOR_IP package contains subprograms for performing some image processing operations on GeoRaster objects. For reference and usage information about the subprograms, see SDO_GEOR_IP Package Reference.

The following other changes are made in this book:

• The previous single chapter about basic GeoRaster operations is divided into two chapters: GeoRaster Database Creation and Management and GeoRaster Data Query and Manipulation.

• Many new topics have been added. These include:

  • GeoRaster Spatial Web Services

  • MapViewer and GeoRaster

  • Using GDAL from the Spatial and Graph Installation

  • Loading with Blocking and Optimal Padding

  • Loading JPEG and JPEG 2000 Images Without Decompression

  • Checking GeoRaster Tables and Objects in the Database

  • Subsetting GeoRaster Objects with Polygon Clipping

  • Performing Cross-Schema Operations

  • Managing Memory to Improve Performance

  • Updating GeoRaster Objects in a Loop

  • Stack Statistical Analysis

  • Logical Operations

  • Image Warping

  • Image Affine Transformation and Scaling

  • Image Stretching, Normalization, Equalization, Histogram Matching, and Dodging

  • Image Filtering

  • Color Balancing During Mosaicking

By default the GeoRaster feature is disabled after the Oracle Spatial and Graph is initially installed. To enable GeoRaster, follow these steps:

1. Connect to the database as SYS AS SYSDBA.

2. Enter the following statement:

   EXECUTE MDSYS.enableGeoRaster;
   

For more information, see Section A.2, "Ensuring That GeoRaster Works Properly After an Installation or Upgrade" in Oracle Spatial and Graph Developer's Guide.

The GeoRaster Java API is updated to support all new packages, functions, procedures and other features in Release 12.1. These include raster algebra, advanced image processing, relational RDT tables, and new metadata. The GeoRaster Java API is included in Oracle Spatial and Graph Java API Reference.

Before you install the LizardTech plug-in, you must follow some new steps to prepare for the installation. For more information, see Third-Party Plug-ins for Compression.

The user interface of the GDAL-based ETL tool is enhanced to improve usability and some new loading options are added. It is also updated with the latest GDAL version. For more information, see the updated user's guide ($ORACLE_HOME/md/demo/georaster/tool/georaster_etl_user_guide.pdf).

The SDO_GEOR_AGGR.mosaicSubset and SDO_GEOR_AGGR.getMosaicSubset procedures accept the new resFilter keyword to the mosaicParam parameter to control the usage of the resolution filter. In addition, HIGHRES is a new commonPointRule keyword option to select the highest resolution image at the overlapping area in the mosaic. These keywords are included in Table 9-1.

The following changes are made in this book:

• The new Section Parallel Compression_ Copying_ Subsetting_ Reprojection_ and Rectification describes how to apply parallel processing to compress, copying, subsetting, and rectification. Information has also been added to Parallel Processing in GeoRaster.

• The new Adding Temporary Tablespaces for GeoRaster Users is added.

The following features are new in this release.

Topics:

• Raster Algebra and Analytics
  
• Advanced Image Processing and Virtual Mosaic
  
• GeoRaster Core Enhancements
  
• Other New Server-Side Subprograms and Features
  
• New and Enhanced Client-Side Tools and Java API
  
• SDO_GEOR_AGGR Package
  
• SDO_GEOR_RA Package
  

The following features are deprecated in this release, and will be desupported in a future release:

• Support for JPEG-B compression in GeoRaster

  You are encouraged to use the JPEG-F support, as explained in JPEG-B Support Deprecated.

The following are additional changes in the release:

• The new Raster Algebra and Analytics describes the raster algebra language (PL/SQL and algebraic expressions) and related raster operations, including conditional queries, cell value-based updates or edits, mathematical operations, classify, on-the-fly statistical analysis, and their applications in cartographic modeling.

• The new Image Processing and Virtual Mosaic describes the advanced image processing capabilities, including advanced georeferencing, reprojection, rectification, orthorectification, raster update, raster appending, large scale physical mosaicking, virtual mosaic, and on-the-fly spatial queries over virtual mosaics.