High-Voltage Disconnector State Identification: synthetic and real images of substation disconnectors

This dataset contains the training and test images used in the paper "Exploring Synthetic Data for Training Deep Learning Models for High-Voltage Disconnector State Identification", currently undergoing peer review.

Contais about 940,000 synthetic (CGI-rendered) and 60,000 real (camera-captured) samples of four types of substation disconnectors, on both open and closed states:

230 kV center break (type 1, as indicated in the article);

230 kV center break (type 2);

230 kV double side break

525 kV horizontal semi-pantograph

Each zip file contains images of one type of substation disconnector. Images are sized 320x128 and are organized in folders, as follows:

00_train_synth: Synthetic training images.

01_train_real: A small set of real training images, as indicated in the article.

02_test_real_normal1: One set of real test images.

03_test_real_normal2: Another set of real test images, from a different time period.

04_test_real_maneuvers: A special set of real test images in which the switches have been operated (are in different states).

As the paper still hasn't been peer-reviewed and published, we have initially set the file visibility as "Restricted". If the paper is published, we are going to open the dataset to the public and fill out the "Publishing information" section.