Sparse group regression and its application in Clustering optimizationand face recognition

The classification of atoms in the sparse representation can be in supervised format and use thetraining tag, or it can be done in unsupervised format based on the clustering of atoms. Themaximum correctness classifier is optimal with defaults and this optimality is in the conditionsthat the probability density function of the data is available, the appropriate density function ofthe data can be estimated when the limitation of the existence of the training data is not infinite.In this article, sparse group regression is first introduced, and then we will discuss how tooptimize and use it in face processing, and finally, we will examine and analyze the performanceof different regressions for face classification.

کلیدواژه ها:

Sparse group regression ، penalty Norm ۲ ، face recognition ، data training