Shulou(Shulou.com)11/24 Report--

Thanks to CTOnews.com netizen Xiao Zhan for the clue delivery! CTOnews.com June 6, the Central Meteorological Observatory announced today that it has initially built a typhoon monitoring and forecasting system based on artificial intelligence. The system can provide important technical support for improving the intelligence of typhoon monitoring and forecasting in China and expanding the global multi-sea tropical cyclone business.

Qian Qifeng, deputy director of the Typhoon and Marine Meteorological Forecast Center of the Central Meteorological Observatory, said, "the Central Meteorological Observatory, in conjunction with various scientific research institutes and universities, has carried out a series of explorations of artificial intelligence in typhoon monitoring and forecasting. In addition, techniques such as typhoon vortex identification, typhoon intelligent strength determination and typhoon rapid enhancement discrimination have been developed, which can give full play to their advantages in dealing with nonlinear and massive data and help forecasters add to the forecast accuracy. "

▲ image source is based on depth satellite image target detection by the Central Meteorological Observatory of China Meteorological Administration. Aiming at the difficulty of weak typhoon vortex recognition, a multi-scale iterative SSD (Single ShotMultiBox Detector) target detection model is proposed. Typhoon vortex coarse positioning and fine positioning are used to intelligently identify typhoon eddies and locate them quickly on infrared cloud images with a large amount of cloud noise (non-typhoon vortex information). The recognition rate of this technology for typhoon and above intensity is close to 100%, and for weak eddies (tropical low pressure level) with no obvious vortex characteristics, the recognition rate can also reach 50%-80%.

According to Zhou Guanbo, chief forecaster of the Typhoon and Marine Meteorological Forecast Center of the Central Meteorological Observatory, through the establishment of typhoon vortex identification model, typhoon intelligent fixed intensity model, typhoon rapid enhancement discrimination model, etc., the Central Meteorological Observatory has initially constructed a typhoon monitoring and forecasting system based on artificial intelligence. It provides important technical support and reference for improving the intelligence of typhoon monitoring and forecasting in China and rapidly expanding the global multi-sea tropical cyclone business.

The Central Meteorological Observatory said that "the Typhoon and Marine Meteorological Forecast Center of the Central Meteorological Observatory will continue to strengthen the application of artificial intelligence in the field of typhoon monitoring and forecasting," to further promote the integration of artificial intelligence technology in typhoon monitoring, forecasting and services, and to provide innovative technical support for precision monitoring and accurate forecasting of typhoons around the world.

IT House noted that in 2019, the Central Meteorological Observatory jointly with Beijing University of posts and Telecommunications proposed an end-to-end visual intelligent typhoon strength determination model, which is based on the pre-training convolution neural network depth learning model, extracts satellite cloud image data to analyze the characteristics related to typhoon intensity, and then constructs a classification model and a similarity-based retrieval model to obtain decision results according to the features. Finally, by combining the identification results of the two models, the intensity, confidence and reference cloud map of the typhoon are obtained. Through the analysis and learning of a large number of samples, this kind of deep learning method can implicitly extract the deep abstract complex features in the image, and it is more and more used to estimate the intensity of typhoon.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.