Quantum genetic algorithm optimized BP neural network for high-resolution remote sensing image classification

The main reason for the low accuracy of traditional BP neural network in the classification of high-resolution remote sensing images is its slow convergence speed and easy to fall into local extremum. In order to achieve the ideal classification effect, the traditional BP neural network needs to be optimized. Quantum Genetic Algorithm (QGA) combines quantum computing and genetic algorithm, which is easier to induce excellent individuals, and solves the premature phenomenon caused by genetic evolution of genetic algorithm (GA) to some extent. In the QGA-BP model combining QGA and BP neural network, the initial weight thresholds are optimized by QGA algorithm to determine the search space, and then the BP neural network is used to determine the optimal solution in the search space, where the genetic operation of QGA uses the quantum rotating gate adjustment strategy. The QGA-BP model is used to classify a high-resolution remote sensing image of GF-2. The experimental results show that compared with other remote sensing image classification methods such as minimum distance method, maximum likelihood method, support vector machine, BP neural network and genetic algorithm optimized BP neural network, QGA-BP model has the highest classification accuracy and the best classification effect, where the overall classification accuracy reaches 0.9502 and Kappa coefficient reaches 0.9113.