Dataset of Smartphone-Captured Images for Classifying Water Droplets on Eyewear Lenses
Description
To aid in the development of machine learning systems for evaluating the state of eyeglass lenses, the ClearLens dataset provides 240 high-resolution photos. Providing a solid basis for developing algorithms that can differentiate between lenses that are dry and those impacted by water droplets is the goal of this collection. Although this type of data is frequently hard to come by it is essential for creating models that function consistently outside of carefully monitored lab environments. All photos were taken with smartphone cameras to guarantee their practical relevance and to represent the conditions of daily use. To include a broad range of lighting and backgrounds, the photos were taken in a number of settings, including outdoor spaces, automobile interiors, and standard indoor rooms. Models that can generalize across various eyewear types are made possible by the dataset's inclusion of a variety of eyewear styles, including sunglasses and daily prescription lenses. A key feature of this dataset is its balanced design, with an equal number of examples for each class to support the training of unbiased models.
Files
Institutions
- Southeast University