A new band selection method for hyperspectral images based on constrained optimization

One of the new techniques in remote sensing is hyperspectral Imagery (HSI). HIS has found many applications in agriculture, environmental science, etc. Due to the large number of spectral bands in HIS, It is difficult and time-consuming to extract information from it. So, the image band selection is an inevitable step. Band selection is done based on the selection of optimum bands in the image based on some pre-defined criteria. In this paper, a new constrained method for band selection is proposed. In the proposed method, the number of bands is considered fixed, and the method must choose the best combination of bands. To do this, another step is added to the classic Genetic Algorithm to satisfy the constraint whilst the optimization problem is done. Experimental results show that the proposed constrained optimization method outperform classic methods in this field in the terms of overall accuracy.