Remote Video Monitoring Home About the RVM System Image Processing Software Tools Processed Video Examples Other Uses of Video Data

Remote Video Monitoring Processed Video Examples Click on any video image on this page for a larger version (~550x410 pixels, ~130k). Processing of the video images returned by the Lake Erie Remote Video Monitoring station begins by digitizing significant features contained within the images. The changing position of the shoreline, the toe of the bluff, and the top of the bluff can be quantified using the image processing software developed for the Remote Video Monitoring system. An example image processing task could begin by locating the (x,y) image coordinates of the shoreline edge, highlighted in white, with the edge detection program lohi from the SunRaster Toolkit. Next the image coordinates of the toe of the bluff are found within the image. The image coordinates of the top of the bluff are also located within the image. The detected edges of the shoreline, bluff-toe, and bluff-top from November 2, 1993 (far left) when compared to those of April 5, 1994 display the changing face of the bluff. The (x,y) image coordinates can be quantified to provide information about the real world. Using the program geomtool, the surveyed ground control points (GCPs) visible within the images can be interactively selected. The geometry between these objects in the real world can then be computed and a best fit solution for the image geometry produced. With knowledge of the image geometry, the location and size of objects and areas of interest can be measured. At left, the (x,y) image coordinate of the edges described above are transformed to real world coordinates and quantified using the SunRaster Toolkit program transform. The inverse is also possible, as portrayed in the image at left, where the surveyed coordinates of the GCPs and shoreline are transformed to image coordinates and highlighted within the image. The same edge detection algorithms described above have also been used to measure ice growth and stability. The edge of the ice as it grows out from the shore can be detected using the SunRaster Toolkit program thresh. The geometry solution found with geomtool can be used to transform the location of every pixel from the oblique image into a map view, called a rectification, using the SunRaster Toolkit program planview. Long term monitoring can provide data such as in the image at left, which depicts the dynamic ice edge position over time. Video data can be used to track ice flow. The images below show two large chunks of ice, laden with sediment, that have been tracked over the course of 24 hours.