Research On Application Of Improved Grasshopper Optimization Algorithmin Image Quality Evaluation

Since the traditional image quality evaluation methods often have the phenomenon that the objective and subjective evaluation results of image quality are far from each other,the evaluation method based on machine learning has received extensive research and attention.The extreme learning machine is one of the commonly used image quality evaluation methods,but the traditional extreme learning machine has problems such as over-fitting,generalization ability,and prediction accuracy that need to be improved.Grasshopper optimization algorithm is an intelligent optimization algorithm proposed in recent years.It has better optimization performance.By introducing Levi's flight improvement,the optimization ability of grasshopper optimization algorithm can be further improved,which can effectively solve the problems existing in traditional extreme learning machine training.In this paper,the improved grasshopper optimization algorithm and extreme learning machine are combined to apply to image quality evaluation.The main work of this thesis is as follows:(1)An improved grasshopper optimization algorithm based on Levi's flight is designed.In order to prevent the grasshopper optimization algorithm from falling into the local optimal solution,the introduction of the Levy flight mechanism can make the original optimization algorithm have a good performance improvement,better jump out of the local optimal solution,and thus get a better global Optimal solution.Experiments show that the improved grasshopper algorithm has better optimization ability than the original algorithm.(2)An image quality evaluation method based on the improved grasshopper optimization algorithm to optimize the extreme learning machine is proposed.A combination of improved grasshopper optimization algorithm and extreme learning machine algorithm is used,and the improved grasshopper optimization algorithm is used to optimize the weights and biases of the extreme learning machine to find the best combination.The test is carried out on the database,and the relevant experimental results show that this method can improve the prediction accuracy in image quality evaluation.In summary,this thesis has carried out research on the improvement of the grasshopper optimization algorithm and the application of the improved grasshopper optimization algorithm optimization extreme learning machine algorithm in image quality evaluation.After experimental analysis and demonstration,this thesis proposes an improved grasshopper optimization algorithm based on the optimization of extreme learning machine has obvious prediction effect on image quality evaluation,meets the expected goal,and conforms to the development trend.