<< Click to Display Table of Contents >>

 

Resample Method

 

 

When the raster layer is overlaid in a different coordinate system or different resolution, the raster layer needs to be resampled. It might need a pixel of the rater data to replace a group of pixels or need a group of pixels to replace a pixel. Therefore, the resampled image might be rougher than the original image.

 

 

 

 

In "Display" tab of Layer Properties window, there are 4 resample methods provided, such as Nearest Neighbor, Bilinear Interpolation, Cubic Convolution, and Majority.

 

Nearest Neighbor: In the process of resampling, the resampled image might correspond to the non-integer when the output image corresponds to the input image. In order to calculate the values of the cells, the value will be calculated from the nearest neighbor cells of the input image. Therefore, the method does not change the value of the cells of the input layer; meanwhile, it is the fastest way to resample and often used for categorical data and thematic data.

 

Bilinear Interpolation: It is a resample method to use four nearest cells to calculate a new value. Since the value is the weighted average of four nearest cells, the image process with the method will be smoother than with "Nearest Neighbor."

 

Cubic Convolution: Similar to "Bilinear Interpolation", "Cubic Convolution" uses the weighted average of nearest 16 cells to calculate a new value. As a result, the output image is calculated by more cells and the image processed with "Cubic Convolution" is clearer than with "Bilinear Interpolation." "Bilinear Interpolation" and "Cubic Convolution" are often used for continuous data, like height, slope, and so on.

 

Majority Interpolation: It is more suitable for the categorized images rather than continuous data. "Majority Interpolation" calculates the values of the pixels in the array and uses the maximum value as the new one.

 

Consequently, you can choose the resample method suitable for you in "Resample Method" drop-down list and click "OK." The resampled image will be displayed in the map window.

 

The following pictures are the samples processed with different resample methods.

 

 

Nearest Neighbor

 

Bilinear Interpolation

 

 

 

Input image

 

 

 Cubic Convolution

 

Majority Interpolation

 

 

.


©2015 Supergeo Technologies Inc. All rights reserved.