圖像品質評估演算法在圖像壓縮品質，或圖像相似度估測上都有廣泛的應用，其中UQI(Universal Image Quality Index)演算法更是在評測圖像相似度的應用上有很好的效果。PSO(Particle Swarm Optimization)是一個很容易實現的演算法，並且具備參數設定少及快速收斂的特性，應用在圖像分割、路徑規劃、圖形辨識及尋求最佳解等都有很好的效果。本研究提出運用PSO結合UQI的方式，並透過調整PSO各項參數以識別頻譜訊號並提升搜尋處理的時效。本論文研究UQI識別頻譜訊號，及PSO結合UQI識別頻譜訊號的方法，並對二者的處理效能進行比較。從研究結果發現，透過適當的參數調整，PSO結合UQI的方式具備更佳的處理速度，可以大幅提昇頻譜訊號識別的處理效能。Image quality assessment algorithms are widely used in image compression quality or image similarity estimation. UQI(Universal Image Quality Index) algorithms has very good results in the evaluation on the image similarity. PSO(Particle swarm optimization) is a very easy to implement algorithms. It has less parameter setting and fast convergence characteristics, and it has good effect in image segmentation, path planning, pattern recognition, and seeking the optimal solution. This study proposes a way which combine PSO and UQI, and through adjustment PSO parameters to get better timeliness of the recognition spectrum signal. This thesis uses UQI algorithm alone and PSO combine UQI to identify spectrum signal, and it compares the processing performance of the two methods. From the research we can understand that just through appropriate parameter adjustments, the method of PSO combined with UQI can have better processing performance for improving the efficacy of the spectrum signal identification.
