Labeled Tesla Vehicle Image Dataset under Varying Lighting Conditions

DescriptionThis dataset contains labeled images of Tesla vehicles captured under varying lighting conditions. The dataset is intended to support research in computer vision, particularly in vehicle classification under realistic and challenging illumination scenarios.Images in this dataset were captured in real-world environments, primarily in public parking lots, parking garages, and along municipal and county roads, in the city of Haugesund, Norway. These environments were deliberately chosen to ensure that the images reflect situations where vehicles are typically encountered in practice.The images were acquired using mobile phones and originally stored in Apple’s HEIC format, which is the default image format on iOS devices. To ensure compatibility with annotation tools, image processing software, and machine learning frameworks, all images were converted to JPG/JPEG format prior to further processing. The conversion was performed without adjusting image quality or applying any post-processing.This dataset covers variations in:vehicle type, model year, and color,camera angle and distance,background and surroundings, andlighting conditions (bright, low-light, and dark). Data Structure The dataset is distributed as four main components:Images:  Tesla-dataset.zipA compressed (.zip) file containing all images in JPG/JPEG format. Each image is identified by a unique filename, which serves as the primary reference across the dataset.Labels:  tesla_dataset_labels.csvA structured label file (CSV format) that provides class annotations for each image. Each row corresponds to a single image and includes the filename along with its associated label(s). This file is intended for use in supervised learning tasks such as image classification and enables straightforward integration with common machine learning frameworks.Labels description labels_description.txt: a text file descriping the labels structure. Annotation configuration: label-studio-code.xmlAn XML file defining the annotation schema used in Label Studio. This file specifies the labeling interface and class structure used during dataset creation, ensuring transparency and reproducibility of the labeling process. It can be used to reload or extend the annotation project within Label Studio.