A novel and efficient algorithm is proposed to reduce the computational complexity for KNN classification.It uses two important features

the approximation coefficient of a fully decomposed feature vector with Haar wavelet and variance of the corresponding untransformed vector

to produce two efficient test conditions.Since those vectors that are impossible to be the

k

closest vectors in the design set are kicked out quickly by these conditions

this algorithm saves largely the classification time and has the same classification performance as that of the exhaust search classification algorithm.Experimental results based on texture image classification will verify our proposed algorithm.