In one embodiment the step of identifying features in the video image data or other sensor data (e.g. for each of the video camera(s) and/or sensors(s)) comprises identifying features in one or more regions of the video image data and/or the other sensor data. In one embodiment, the regions of the video image data and/or other sensor data in which features are identified comprise blocks of data. Thus preferably the video image data and/or other sensor data is divided into blocks for the purposes of comparing the video image data and/or other sensor data. The blocks of data preferably comprise square arrays of data elements, e.g. 32×32 or 64×64 pixels (although any suitable and desired shape and size of blocks may be used). Identifying features in regions (e.g. blocks) of the video image data or other sensor data helps to simplify the processing task of identifying such features by reducing the area over which features are identified (and thus the amount of data that has to be processed).

The step of identifying features in the video image data or other sensor data is preferably performed individually for (e.g. each of) the video camera(s) and/or sensors(s). In one embodiment, once this has been performed, the same or similar features that have been identified in the video image data or other sensor data from the plurality of video camera(s) and/or sensors(s) are matched to each other.