As defined above, f is map matching function which maps ordered time stamped sequence of location data traces to edges of the road network.
The reviewed approaches are classified according to their nature (area-based and feature-based) and according to four basic steps of image registration procedure: feature detection, feature matching, mapping function design, and image transformation and resampling.
Overall, normalised matching mapped 3,266 (44.08%) terms, Expanded String Matching further mapped 304 (4.10%) terms, WordNet Matching give another 92 (1.24%) newly mapped terms, and Substring Matching increased by 809 (10.88%) the matched terms.
Given the large percentage of identical and/or highly matching mapping data, a library of PS26_EST_OTHERS was also established using the same parameters as BC8_EST_OTHERS.
For this reason, every matched distance (between GPS point and matched map position) should fall below a given threshold (fixed at 20 m, after several experiments).
The reason for weighting by the neighborhood kernel is to avoid rigid overlaying which could result from focusing on the best-matching map nodes only; there may exist several closest best-matching nodes for an input data vector.
To score the generated isotopic match map, we use three different metrics.
M j) denotes the optimal match map, or sequence of matched isotopic distributions up to distribution j.
We then can build the M j) matrix from Equations (1– 7) to yield the isotopic distribution match map.
