A new fast kNN classification algorithm is presented for texture and p
attern recognition. The algorithm identifies the first k closest vecto
rs in the design set of a kNN classifier for each input vector by perf
orming the partial distance search in the wavelet domain. Simulation r
esults show that,without increasing the classification error rate, the
algorithm requires only 12.94% of the computational time of the origi
nal kNN technique.