Enhancement Filters - MapInfo_Pro_Advanced - 2023

The types of enhancement filters include:

• Sharpening filters - Sharpening filters tend to retain high frequency information within a grid while reducing the low frequency information. The kernel of the sharpening filter is designed to increase the brightness of the center cell relative to its neighboring cells. The kernel array usually contains a single positive value at its center, which is surrounded by negative values. The following is an example of a 3x3 kernel for a sharpening filter.

  				-1.0	-1.0	-1.0
  				-1.0	 9.0	-1.0
  				-1.0	-1.0	-1.0
  			

  The sharpening filters available in MapInfo Pro Advanced are:

  1. 3x3 Edge Sharpen
  2. 5x5 Edge Sharpen
• Line and Edge Enhancing filters - Line and edge enhance filters enhance image features with specific directional components. An edge enhancement filter sharpens edges between objects. For example, when a water body meets a population habitat. In a line or edge enhancing kernel, the x-values are weighted such that low frequency variations are removed and the boundary between multiple regions are highlighted. The following is an example of a 3x3 kernel for an edge enhancement filter.

  				-1.0  -1.0  -1.0
  				 2.0   2.9   2.0
  				-1.0  -1.0  -1.0
  			

  The line and edge enhancement filters available in MapInfo Pro Advanced are:

  1. 5x5 Horizontal Edge Enhance
  2. 3x3 Vertical Edge Enhance
  3. 3x3 Horizontal Line Enhance
  4. 3x3 Vertical Line Enhance
• Laplacian filters - A Laplacian filter enhances the edges of a grid. This filter convolves a kernel of weights against each cell in the grid and its neighboring cells. The kernel usually has a high central value, surrounded by negative central values. The following is an example of a 3x3 Laplacian kernel.

  				0	-1 	 0
  				-1	 4 	-1
  				0	-1	 0
  			

  The Laplacian filters available in MapInfo Pro Advanced are:

  1. 3x3 Laplacian
  2. 9x9 Laplacian
• Laplacian of Gaussian filter - Laplacian filters are highly sensitive to noise. Therefore, it is sometimes a common practice to smooth the image using a Gaussian filter before applying a Laplacian filter. This two-step process is call the Laplacian of Gaussian (LoG) operation. In any grid, applying LoG filter results in faster computation times on large datasets, as smoothing of the grid is already achieved using a Gaussian filter. In this method, areas on a grid that have a gradient of zero will produce values of zero in the laplacian filtered image, while areas with high gradient values will produce positive or negative values in the image. The LoG filter then enhances the contrast in the original image and makes the edges sharper. The following is an example of a 9x9 Laplacian of Gaussian kernel.

  				0	1	1	2	2	2	1	1	0
  				1	2	4	5	5	5	4	2	1
  				1	4	5	3	0	3	5	4	1
  				2	5	3	-12	-24	-12	3	2	1
  				2	5	0	-24	-40	-24	0	5	2
  				2	5	3	-12	-24	-12	3	2	1
  				1	4	5	3	0	3	5	4	1
  				1	2	4	5	5	5	4	2	1
  				0	1	1	2	2	2	1	1	0
  			

  The LoG filter available in MapInfo Pro Advanced is 9x9 Laplacian of Gaussian.

• Sobel filters - A Sobel filter is used to detect edges by applying a horizontal and vertical filter in sequence. When a vertical kernel is used, the difference of pixel intensities in an edge region is calculated. The mask calculates the difference of right and left pixel values around that edge. This gives more weightage to cell pixel values around the edge which increases the edge intensity of the filtered image. An example of vertical kernel is shown below:

  				-1	0	1
  				-2	0	2
  				-1	0	1
  			

  A horizontal kernel finds edges in a horizontal direction. When you convolve this mask onto a grid, the horizontal edges in the filtered image become prominent. An example of horizontal kernel is shown below:

  				1	2	1
  				0	0	0
  			   -1   -2   -1
  			

• Roberts cross filters - Roberts cross filters highlight regions of high spatial frequency in the grid. The Roberts filters provided in MapInfo Pro Advanced are designed to produce maximum responses over edges running at 45^o to the pixel grid. One filter kernel is provided for both perpendicular orientations, G_x and G_y. The filters are applied to calculate the gradient in each orientation or find the magnitude and orientation of each individual gradient. An example of the two kernels are shown below:

  				1	0				
  				0	-1
  			

  				0	1
  			   -1	0
  			

  The Roberts cross filters available in MapInfo Pro Advanced are:

  1. Roberts cross Gx
  2. Roberts cross Gy