Kerala Vehicle Colour Classification Dataset
收藏资源简介：
Kerala Vehicle Colour Classification Dataset (KVCC) Description This dataset consists of 1,104 high-resolution vehicle images captured from ANPR camera feeds in the Kozhikode district of Kerala, India with support from the Kerala Police Department. The dataset was curated to support research in vehicle attribute recognition under unconstrained environmental conditions, specifically addressing challenges like sunlight glare and varied illumination. Dataset Structure The images are organized into 5 primary color classes: Black Blue Gray Red White The data is split into Training (80%), Validation (10%), and Testing (10%) sets. Naming Convention All images have been renamed using the format: KVCC_[Index].jpg Metadata The included 'metadata.csv' provides a mapping of: Filename: The unique KVCC ID. Label: The ground truth color class. Usage This dataset is intended for academic and research purposes. If you use this dataset, please cite: Arun Natarajan et al.Robust Vehicle Colour and License Plate Recognition in Indian Traffic Conditions: A Hybrid Approach using YOLO and Multi-View Consensus Scoring. Authors Arun Natarajan, Kamble Aditya Dattatray, Thareesh Prabakaran, Rachel Aji Joseph, Abhinav Kurup, Hafeez Muhammed, Shihabudheen K. V., Munavar Fairooz C. National Institute of Technology Calicut, Kozhikode, Kerala, India License: Creative Commons Attribution 4.0 International (CC BY 4.0)
喀拉拉邦车辆颜色分类数据集（Kerala Vehicle Colour Classification Dataset，简称KVCC） 数据集描述 本数据集由印度喀拉拉邦科泽科德地区的自动车牌识别（Automatic Number Plate Recognition，ANPR）摄像头采集的1104张高分辨率车辆图像构成，采集工作得到了喀拉拉邦警察厅的支持。本数据集旨在支持非受限环境下的车辆属性识别研究，专门针对阳光眩光、光照条件多变等典型挑战进行采集构建。 数据集结构 图像共分为5个主要颜色类别：黑色、蓝色、灰色、红色、白色。 数据集划分为训练集（80%）、验证集（10%）与测试集（10%）。 命名规范 所有图像均采用KVCC_[序号].jpg的格式进行重命名。 元数据 随附的metadata.csv文件包含如下映射关系： - 文件名：唯一的KVCC标识符 - 标签：真实颜色类别（标注真值） 使用说明 本数据集仅用于学术与研究用途。若使用本数据集，请引用如下文献： Arun Natarajan 等. 《印度交通场景下鲁棒的车辆颜色与车牌识别：基于YOLO与多视图共识评分的混合方法》 作者信息 作者：Arun Natarajan、Kamble Aditya Dattatray、Thareesh Prabakaran、Rachel Aji Joseph、Abhinav Kurup、Hafeez Muhammed、Shihabudheen K. V.、Munavar Fairooz C. 单位：印度喀拉拉邦科泽科德市国立理工学院 许可证 知识共享署名4.0国际通用许可协议（Creative Commons Attribution 4.0 International，CC BY 4.0）
