Solar cell data image

How are solar cell images annotated?

Every image is annotated with a defect probability (a floating point value between 0 and 1) and the type of the solar module (either mono- or polycrystalline) the solar cell image was originally extracted from. The individual images are stored in the images directory and the corresponding annotations in labels.csv.

Are defective solar cells affecting the power efficiency of solar modules?

The dataset contains 2,624 samples of 300x300 pixels 8-bit grayscale images of functional and defective solar cells with varying degree of degradations extracted from 44 different solar modules. The defects in the annotated images are either of intrinsic or extrinsic type and are known to reduce the power efficiency of solar modules.

What is an example image chip in a solar panel dataset?

An example image chip would be “solarpanels_hd_1__x0_0_y0_14027_dxdy_832.tif,” which is an HD chip from Tile 1 in the solar panel dataset, with a minimum x bound of 0, minimum y bound of 14,207 and window size of 832 by 832 pixels. The corresponding label file for this example would be “solarpanels_hd_1__x0_0_y0_14027_dxdy_832.txt.”

Which datasets include annotated solar panels in native resolution and HD satellite imagery?

The complete dataset contains native resolution satellite imagery, corresponding HD imagery, and solar panel object labels for each image type (Fig. 1). To the best knowledge of the authors, there are no publicly available datasets including annotated solar panels in native resolution and HD satellite imagery.

What is a benchmark dataset for semantic segmentation of solar cell images?

A benchmark dataset for semantic segmentation of solar cell images is published. Twelve deep learning models are trained for defect detection in EL images. Benchmark IoU and recall metrics are provided for 5 of the 24 labelled classes.

Can deep learning detect defects in crystalline silicon solar cells?

This paper presents a benchmark dataset and results for automatic detection and classification using deep learning models trained on 24 defects and features in EL images of crystalline silicon solar cells. The dataset consists of 593 cell images with ground truth masks corresponding to the pixel-level labels for each feature and defect.