Arti cial neural networks trained on spectral and textural features extracted from Advanced Very High Resolution Radiometer (AVHRR) images have been used to develop an automated cloud classi cation system. Selection of the optimum combination of features was achieved by using statistical methods presented in earlier work by Gu et al. and by running large numbers of neural network simulations on test datasets. The performance of these methods surpasses that of other approaches such as the use of Gabor lters for texture segmentation and the maximum likelihood classi er. A particular architecture for an operational classi cation system is presented based on a two-stage multiple network con guration which is shown to segment complex images to a high degree of accuracy and achieves an overall accuracy on an independent, representative test set of 91%.
Artificial neural networks have been presented with a set of spectral and textural features extracted from Meteosat images and their performance in distinguishing a number of cloud types has been compared to more traditional techniques. It is shown that back-propagation neural networks can perform extremely well, reaching high classification accuracies (of order 95%) on this type of data. A maximum likelihood classifier achieves a highest possible success rate of only 90% in comparison. A competitive learning network fares considerably less favourably and is probably not well suited to this particular data set. Neural networks in general are a powerful interpretative tool, allowing us to examine key features of the original data by exploring how the networks respond and learn.
In Scotland all 13 universities and a number (8) of other related institutions of higher education are connected on a high bandwidth computer network. Organizationally, this is composed of four interconnected Metropolitan Area Networks (MANs). The high bandwidth (155 Mbit/s) and the network's operation under ATM (Asynchronous Transfer Mode) enables routine use of high quality video-conferencing between institutions. Each institution has at least one (all the universities have two) video-conferencing suites, fully equipped with audio and video equipment and direct network access for PCs
Video-conferencing was used to share a short series of lectures between several universities. A high bandwidth network (155Mbit/s) permitted near broadcast TV quality video to be combined with fully mixed, high-quality audio. The lectures were supported by visual aids made available using Microsoft NetMeeting to provide multipoint, shared applications. NetMeeting is shown to be a stable and effective platform for distributing multimedia material at a much higher resolution than is possible using the video signals common in most video-conference lectures, although care must be taken when constructing animated material.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.