Algorithm-Oriented SIMD Computer Mathematical Model and Its Application

摘要

This paper has designed a professional and practical SIMD computer mathematical model based on the SIMD physical machine model combined with the variable addition method. Furthermore, the model is applied in image collection, processing, and display operations, and a SIMD data parallel image processing system is finally established by absorbing the parallel computing advantages of the mathematical model. In addition, the data-parallel image processing algorithm is introduced and the convolutional neural network algorithm is optimized to promote the significant improvement of the main performance such as the accuracy of the application system. The final experimental results have shown that the highest accuracy of the data-parallel image processing algorithm reaches 93.3% and the lowest error rate reaches 0.11%, which proves the superiority of the SIMD computer mathematical model in image processing applications.