Performance comparison of edge detection algorithms for satellite images using bigdata platform spark

摘要

Edge Detection has been a vital step in the field of image processing. It is used to identify the breaks in depth, surface orientation or changes in the brightness. It is one of the most vital steps for object detection in the image. In today's big data world, it becomes crucial to know the different algorithms that can run faster in order to deal with the fast generating data. To process the large volume of data big data platforms like Hadoop are being leveraged. In this paper we have implemented different edge detection algorithms such as Canny Edge, Sobel operator and Laplacian of Gaussian on Apache Spark. We have executed our implementations on remote sensing images residing on HDFS and then compared the statistical performance and scalability of all these algorithms for different set of images.