부산대학교 도서관

Cloud-to-ground separation is a basic work in the application of satellite cloud image data. By analyzing color satellite images, a cloud-to-ground separation algorithm based on pixel color and gray value is proposed, which achieves cloud-to-ground separation. In this method, the difference of image RGB component and gray value are used as discriminant conditions, and the cloud area is recognized by dynamic threshold method. The bigger the difference of RGB components and the smaller the gray value of the image, the larger the range of the cloud area reserved. Histogram area similarity measure is used to determine the recognition process. The experimental results show that this method can effectively reduce the occurrence of false alarm and achieve better recognition results than the fixed threshold.