DVRL-VST-based electric smelting furnace working condition video identification method
The invention discloses a DVRL-VST-based working condition video identification method for an electric smelting furnace for magnesia. The method comprises the following steps: 1, carrying out abnormal sample augmentation through a CycleGAN style migration architecture; 2, inputting the enhanced data...
Saved in:
Main Authors | , , , , , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
02.08.2022
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a DVRL-VST-based working condition video identification method for an electric smelting furnace for magnesia. The method comprises the following steps: 1, carrying out abnormal sample augmentation through a CycleGAN style migration architecture; 2, inputting the enhanced data into a DVRL-VST network based on reinforcement learning sample value evaluation and VST for training; and 3, performing working condition identification on the operation video of the electric smelting magnesia furnace by utilizing a training result. According to the method, interference caused by environment light changes of a production site and changes of inherent visual features of different electric smelting furnace shells of the magnesium smelting furnace can be eliminated, video signals of the working conditions are used, the spatio-temporal features of local gradual change of the abnormal working conditions of the electric smelting furnace of the magnesium smelting furnace are extracted from the two dimensi |
---|---|
Bibliography: | Application Number: CN202210474871 |