Is there any difference between blobs and connected components?
As I understand it the term “connected component labelling” refers to the final step in the process of "blob detection”, whereby each separate group of connected pixels in a binary image (for example a thresholded background difference image) is identified.
In contrast, the term “blob detection” can either be used to mean the same thing as “connected component labelling” or to refer to the two step procedure of generating a binary image AND performing connected component labelling.
The resulting connected groups of pixels can nonetheless be called “blobs" or "connected components” interchangeably.
It is also worth mentioning that "connected components labelling” has a more general meaning in terms of graph theory where “connected components” are groups of nodes where there exists a path joining any two members of that group. Connected groups of pixels in a binary image thus correspond to a special case of this broader definition, where pixels can be regarded as nodes with arcs to those pixels in their 4- or 8- pixel neighbourhood that share the same value. Hope this helps!