Optimization of Remote Sensing Image Segmentation by a Customized Parallel Sine Cosine Algorithm Based on the Taguchi Method

Affected by solar radiation, atmospheric windows, radiation aberrations, and other air sky environmental factors, remote sensing images usually contain a large amount of noise suffer from problems such as non-uniform image feature density. These bring great difficulties to the segmentation high-precision image. To improve effect images, this study adopted an improved metaheuristic algorithm optimize parameter settings pulse-coupled neural networks (PCNNs). Using Taguchi method, optimal parallelism scheme was effectively tailored for specific target problem. The blindness in design parallel structure avoided. superiority customized SCA based on method (TPSCA) demonstrated tests with different types benchmark functions. In study, simulations were performed using IKONOS, GeoEye-1, WorldView-2 satellite images. results showed that accuracy proposed model significantly improved.