Component-based global k-NN classifier for small sample size problems

The classical k-NN classifier has been widely used in pattern recognition. However, it does not take into account the structural information of local samples. This paper presents a novel classifier named component-based global k-NN classifier (CG-k-NN), which takes advantage of the structural information of the local neighbors for enhancing the classification performance. We choose k nearest neighbors of a given testing sample globally at first, and then use these neighbors to represent the testing sample via ridge regression. In the further step, we construct the component image of each class by using the intra-class images from the k nearest neighbors and the corresponding representation coefficients. Finally, the testing sample is assigned to the class that minimizes reconstruction residual. The proposed method CG-k-NN is evaluated using the ORL, FERET, AR face image database and PolyU palmprint databases. The experiment results demonstrate that our method is more efficient and effective than the state-of-the-art methods such as sparse representation based classifier (SRC) and linear regression based classifier (LRC).

► A novel classifier named component-based global k-NN classifier (CG-k-NN) is studied. ► It takes advantage of the structural information of the local neighbors for enhancing the classification performance. ► We construct the component image of each class by using intra-class images from the local neighbors. ► The experiment results demonstrate that our method is more efficient and effective than the state-of-the art methods.