I am going to create raster from points and I want the raster to calculate the mean value of z-value of the points that are located in cell size and put it as cell value in the raster.
I used NumericRasterizer but it does not describe how it calculates pixel values of the raster.
Any Ideas?
The general rule of thumb dealing with rasters is last in wins. So if you have multiple features in one raster cell, the output will be the value of the last feature to arrive at the NumericRasterizer.
My first thought is send your points to a Tiler set to the same number of rows and columns as your output raster. Use an Aggregator grouped by the tiler row and column attributes with the Attributes to average set to your Z values. Then use a CenterPointReplacer to turn your aggregate into a single point with the mean value of all the points in the cell.
There is probably also a way to turn your points into a uniform point cloud and work that way, but I'm not as familiar with those transformers