Prediction method and system for national ecological environment information visualization based on PredRNN

The invention discloses a PredRNN-based nationwide ecological environment information visualization prediction method and system, and the method comprises the steps: embedding a PredRNN deep learning model in a prediction function of a visualization system, introducing the induction deviation of gro...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: SHI HAOYUAN, TU LIJING, BIAN JINGE, ZHANG YANCONG, TAO ZHENG, ZHAO DEHAN, WU GUODONG, ZHANG NENGXIANG
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses a PredRNN-based nationwide ecological environment information visualization prediction method and system, and the method comprises the steps: embedding a PredRNN deep learning model in a prediction function of a visualization system, introducing the induction deviation of group invariance in space into space-time prediction through employing a convolution layer, and carrying out the prediction of the prediction function. The method has higher modeling capability and higher calculation efficiency for the historical observation sequence. Real-time ecological environment data are obtained through a web crawler technology, the data are processed and then enter a PredRNN prediction model for prediction, a visualization tool is called through static resources, a main page is constructed, service and request are responded based on a cloud server side, and the data are obtained and visually displayed. And finally obtained visual data can be used for query of the majority of users. 本发明公开了一种基于Pr