A Novel Binary QUasi-Affine TRansformation Evolutionary (QUATRE) Algorithm

QUasi-Affine TRansformation Evolutionary (QUATRE) algorithm generalized differential evolution (DE) to matrix form. QUATRE was originally designed for a continuous search space, but many practical applications are binary optimization problems. Therefore, we novel version of QUATRE. The proposed is implemented using two different approaches. In the first approach, new individuals produced by mutation and crossover operation binarized. second binarization done after mutation, then cross with other performed. Transfer functions critical binarization, so four families transfer introduced algorithm. Then, analysis performed an improved function proposed. Furthermore, in order balance exploration exploitation, liner increment scale factor Experiments on 23 benchmark show that approaches superior state-of-the-art algorithms. Moreover, applied it dimensionality reduction hyperspectral image (HSI) test ability solve experimental results HSI imply methods better than Principal Component Analysis (PCA) Linear Discriminant (LDA).