MapInfo Pro 17.0 and Warp Image functionality

Product affected: MapInfo Pro Advanced™
MapInfo Pro Advanced 17.0 introduces the "Warp Image" capability. 
On the RASTER tab, scroll down to the Utilities section in Raster Operations, to find the "Warp Image" tool:

User-added image

What is Warping?
Warping is very similar to "georeferencing", i.e. transforming a raster dataset using the source (pixel coordinates) and target (world coordinates) control points. The process maps a series of known X and Y coordinates with the points on the raster dataset or with locations in the target data.
Control Points
A series of ground control points are required to georeference a raster data. Control points are the locations that can be accurately identified on a raster dataset as well as in real-world coordinates. Control points are a list of image coordinates and their corresponding world coordinates. Control points help us to map image pixels coordinates to world coordinates.

Control points can be created by:
  • Image Pixel Coordinates: The pixel Coordinates of raster Image to be warped.
  • World Coordinates: The world coordinates to which the source raster will be warped. The world coordinates corresponding to the Source image pixel that have been defined/used, and need to be provided.
Modes of Output
The Warp operation produces two possible outputs:
  1. Warped raster image
  2. MapInfo Virtual Raster (MVR) Output
When creating a warped raster image, the output is generated by warping the base resolution level source data. 
With this method, a warped raster output file is typically created on disc.

When creating an MVR, the warped output is generated on the fly from the most appropriate data resolution level. MVR also requires an underview interpolation method to be defined. The MVR output can be displayed immediately in MapInfo Pro and will show the effect of the warp transform.

NOTE : It is recommended to produce an MVR before producing a typical warped raster, as MVR has the obvious advantages of using no disk space, negligible time for generation and can be easily modifiable at a future time.

Supported Transform Types
The User has the option to choose from multiple transform methods available:
  1. Affine
  2. Conformal
  3. Projective
  4. 2nd order Polynomial
  5. 3rd order Polynomial
NOTE:  There is also an "Auto" method which selects the transform that produces the lowest standard deviation error and satisfies the minimum number of points requirements of the method.

Affine is a linear transformation with separate scaling, rotation and shift along the X and Y axes. This can be used when users need to adjust from an unknown coordinate system such as a "local Mine Grid" to UTM. A minimum of three control points are required for this transformation
Conformal transformations preserve shapes and angles and may include a rotation, a scaling and a translation. Straight lines and parallel lines remain straight and parallel in the output image. A minimum of two control points are required for a conformal transformation.
Projective transformations maps lines to lines. Straight lines remain straight but parallelism may not be preserved. Four control points are required for a projective transformation.
2nd Order polynomial needs a minimum of nine control points and transformations are higher-order, non-linear transformations which can handle more complex local distortions. Polynomial transformations are commonly used for image registration and correction of distortions in remote sensing applications. Polynomial transformations are smooth and are also known as 'rubber-sheet' transformations as they enable parts of an image to be stretched or warped to fit the control points.
3rd Order Polynomial is same as 2nd order polynomial but it requires a minimum of sixteen control points.
The minimum number of GCP's (Ground Control Points) required for each method are:
2nd order Polynomial                                      9
3rd order Polynomial                                      16
Use Case
A GIS professional sometimes needs to add geographic information to an image (which can be an air photo, a scanned geologic map, or a picture of a topographic map) along with mapping/correcting it's pixel coordinates to its real world geographical coordinates so that the image can be placed in it's appropriate real world location. In other words apart from
georeferencing, warp also corrects the distorted portions of the image.

Warp tool can be useful when users want to find out where an image of a location taken from a flying plane will lie exactly on a world map. GIS professionals will use a combination of control points, transform types, projections to warp the image and produce the desired output.

Sample Example

Here is an example of Image Warping:
User-added image

This is an Input Image of topographic Map. Geographic information is added to it and mapped with the pixel coordinates to its real world coordinates using 3rd order Polynomial transform in Warp tool, so that the image can be placed in its appropriate real world location.

User-added image

This is an output Warped Image. Upon opening the warped output in a base map, it is clearly seen that India's Map is correctly placed in real world location.

(Thanks to Shweta Shukla from PB who posted this great article in the PB Knowledge Community.
Also thanks to Smriti Pandey)

UPDATED:  April 18, 2019