Spatial aggregation produces images of coarser resolution (grouping pixels in a grid of coarser resolution and calculating mean values) or of coarser scale (by filtering with low-pass filters). Thereby it is a form of generalization that may improve classification results. Spatial aggregation can be applied after classification to get rid of the salt-and-pepper effect.
According to oxford dictionary, "aggregate (noun)" is defined as a whole formed by combining several separate elements. "to aggregate (verb)" is "to form or group into a class or cluster"
Aggregation to
Here (https://www.mdpi.com/2072-4292/10/8/1260/htm) "data aggregation" is used for: Data reduction steps without loss of information are a scientific and computational challenge but are critical to enable effective data processing and information delineation in data-rich studies. We investigated the effect of four spatial aggregation schemes on Landsat imagery
Similarity to [AM14] Generalization and aggregation (but, in addition to scale, involving "resolution")
Similarity to [IP3-6] Neighbourhood analysis (convolution)
Similarity to [IP3-9] Scale space analysis
Apply spatial aggregation for generalizing an image classification
New