A national-scale land cover reference dataset from local crowdsourcing initiatives in Indonesia

This collection represents geographically diverse, temporally consistent, and nationally relevant land cover (LC) reference data collected by visual interpretation of very high spatial resolution imagery, in a national-scale crowdsourcing campaign (targeting seven generic LC classes) and a series of expert workshops (targeting seventeen detailed LC classes) in Indonesia. The interpreters were local citizen scientists (crowd/non-experts) and local LC visual interpretation experts from different regions in the country. This helps to ensure that the LC map products are relevant and can contribute effectively to the actionable information needs of the national and sub-national stakeholders and end users of the LC products within the country. We provide the raw LC reference dataset, as well as a quality-filtered dataset, along with the quality assessment indicators. The dataset is relevant for the LC mapping community, i.e., researchers and practitioners, as reference data for training ML algorithms and for map accuracy assessment (with appropriate quality-filters applied). The dataset is also useful for the citizen science community, i.e., as a sizable empirical dataset to investigate the potential and limitations of the crowd/non-experts, demonstrated for LC mapping in Indonesia for the first time to our knowledge, within the context of complementing traditional data collection by expert interpreters. The detail description of the data and the data collection methodology can be found in our paper below. <br> Hadi, Yowargana, P., Zulkarnain, M.T. <em>et al.</em> A national-scale land cover reference dataset from local crowdsourcing initiatives in Indonesia. <em>Sci Data</em> <strong>9</strong>, 574 (2022). https://doi.org/10.1038/s41597-022-01689-5

本数据集为印尼境内通过目视解译超高空间分辨率影像获取的土地覆盖（Land Cover, LC）参考数据，兼具地理分布广泛性、时间序列一致性与国家适配性。数据采集涵盖两类工作：一是面向7个通用土地覆盖类别的全国性众包项目，二是针对17个细分土地覆盖类别的多场专家研讨会。
参与本次解译工作的人员包括来自印尼各地区的本地公民科学家（众包人员/非专业人士）以及本地土地覆盖目视解译专家，此举可保障土地覆盖制图产品贴合本土实际，有效满足印尼国内国家级、地方级利益相关方以及土地覆盖产品终端用户的实用信息需求。
本项目同步提供原始土地覆盖参考数据集、经质量过滤后的数据集，以及配套的质量评估指标。
本数据集可服务于土地覆盖制图领域的科研人员与从业者，作为训练机器学习（Machine Learning, ML）算法以及开展地图精度评估（需应用适配的质量过滤规则）的参考数据。
此外，本数据集也可用于公民科学领域研究：作为规模可观的实证数据集，用于探究众包人员与非专业人士在土地覆盖解译工作中的潜力与局限——据我们所知，本研究首次在印尼土地覆盖制图场景中验证了该模式对专家解译传统数据采集工作的补充价值。
有关本数据集及数据采集方法的详细说明，请参阅以下论文：
Hadi、Yowargana P.、Zulkarnain M.T. 等. 源自印尼本地众包项目的国家级土地覆盖参考数据集[J]. 科学数据(Sci Data), 2022, 9:574. https://doi.org/10.1038/s41597-022-01689-5