邁入21世紀後國際石油短缺、能源缺乏,促使全球對永續能源發展重視。此外, 隨著永續能源的發展之下,太陽能發電使得在永續能源方面成為發展趨勢。然而, 大陽能板大多設置在戶外,經過風吹日曬雨淋太陽能板上難免就會出現髒汙或遮蔽物,這時候就必須清除太陽能板上的髒汙以避免遮陰效應造成太陽能的轉換效率下降。針對此問題本論文先是利用熱影像提出太陽能板異常偵測演算法,利用熱影像特性可以正確判斷出異常區域的位置以及面積。再來我們參考論文[13],在視覺影像中也能有效的偵測出太陽能板上是否有遮陰異常的情況發生。 除了能夠偵測出異常區域外也要知道異常區域的位置,才能更有效地排除異常,所以我們又提出了一個能將熱像儀影像與視覺影像融合的方法,融合之後的影像能夠更準確的判斷太陽能板與異常區域的相對位置。 經實驗結果證明,使用本論文熱影像的太陽能板異常偵測演算法準確率可達90.623%,與其他方法相比準確度有些微提高。除此之外,本論文也解決了在不同影像尺寸狀況之下的影像融合。
After the 21st century, oil and energy shortage male to pay attention to sustainable energy development around the world. In addition,the solar power have maed a trend in sustainable energy. However, most of the solar panels are installed outdoors. The dirt and shelter will appear on the solar panels. At this time, it must to remove the dirt on the solar panels, that to avoid the solar energy conversion caused by the shadow effect. In order to this problem, this paper proposes a detection algorithm for solar panels in thermal images. It can detect correct position and area. And than we reference [13]. Proposing a algorithm to detect defect for solar panels in visible images. After two algorithm, We have also proposed a method to fuse thermal image with visible image. The fusion image can more accurately detect the position and area of the solar panel. The experimental results show that the accuracy of the thermal image detection algorithm can reach 90.623%. It better than other papers. In addition, this paper also solves image fusion under different image size.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。