Deep Eyes solar Panels

What is deepsolareye?

The Solar Panel Soiling Image Dataset known as DeepSolarEye , comprising 45,754 images of solar panels with power loss labels, was instrumental in our analysis. Our experimental setup included two identical solar panels placed side by side, with an RGB camera facing them.

Can deep learning be used for solar PV image analysis?

In this paper, we presented a first deep learning based application for solar PV image analysis. We proposed a novel fully convolutional neural network based approach which takes an RGB image of solar panel and environmental factors (optional) as inputs to predict power loss, soiling localization, and soiling category.

How to train a deep solar eye model?

A simplified overview of DeepSolarEye is presented in Figure 2. Ideally, to train such a model, we need two kinds of labels: 1) power loss, and 2) localization mask. In our approach, DeepSolarEye, we bypass the explicitly labeled localization data requirement by using power loss label as weak supervision for localization.

Does deepsolareye predict power loss and soiling area simultaneously?

In this paper, we present a novel end-to-end fully con-volutional neural network, DeepSolarEye, that simultaneously predicts the power loss and localizes soiling area from the RGB Fig. 2: Overview of our method, DeepSolarEye, that predicts impact on the power loss and the soiling area simultaneously. image of solar panel.