GIS multi-source partial discharge mode identification method and device, equipment and storage medium

The invention provides a GIS multi-source partial discharge mode recognition method and device, equipment and a storage medium, and belongs to the field of computer vision, and the method comprises the steps: building a backbone network of a YOLOv5 model based on a ShuffleNetV2 network, recognizing...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: HUANG YONGLIANG, LI JUNBIN, YU XINQIU, YANG KAI, CHEN QI, YI XIAOBO, LI WENLONG, JIANG WEI, WU YOUYU, WEN ZHAOLONG
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 provides a GIS multi-source partial discharge mode recognition method and device, equipment and a storage medium, and belongs to the field of computer vision, and the method comprises the steps: building a backbone network of a YOLOv5 model based on a ShuffleNetV2 network, recognizing a plurality of target modules from a large number of ShuffleNetV2 modules in the network based on the number and arrangement mode of shuffling units, and carrying out the recognition of the ShuffleNetV2 modules in the network, the redundant first convolution layer with the convolution kernel of 1 * 1 in the target module is removed through branch replacement, and the calculation amount of convolution calculation is reduced; according to the method, the to-be-detected PRPD atlas is input into the YOLOv5 model, the number of channels of an extracted module mixed feature map is small due to the fact that a first convolutional layer is reduced, the attention degree of high-dimensional semantic information is increased