Gas emission rate inversion method based on deep neural network
The invention provides an SO2 gas emission rate inversion method based on a deep neural network. Comprising the following steps: preprocessing acquired original image data; acquiring an optical flow according to the preprocessed data; dividing the processed data into a training set and a debugging s...
Gespeichert in:
Hauptverfasser: | , , , , , |
---|---|
Format: | Patent |
Sprache: | chi ; eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The invention provides an SO2 gas emission rate inversion method based on a deep neural network. Comprising the following steps: preprocessing acquired original image data; acquiring an optical flow according to the preprocessed data; dividing the processed data into a training set and a debugging set; constructing a deep neural network model and optimizing the deep neural network model; and realizing inversion of the discharge rate by using the optimized deep neural network model. Compared with the prior art, the method has the advantages that the anti-interference capability is high, the problem of image edge plume discharge rate inversion can be solved to a great extent, and more importantly, real-time, rapid and accurate inversion of the plume discharge rate is realized.
本发明提出了一种基于深度神经网络的SO2气体排放速率反演方法。包括以下步骤:对获取的原始图像数据进行预处理;根据预处理后的数据获取光流;将经过处理的数据分成训练集和调试集;构建深度神经网络模型并对其进行优化;利用优化的深度神经网络模型实现排放速率的反演。与现有技术相比,本发明抗干扰能力强,能够极大程度上解决图像边缘羽流排放速率反演的难题,更为重要的是实现了对羽流排放速率的实时、快速、精确地反演。 |
---|