Global Land Cover Assessment Using Spatial Uniformity Validation Dataset

The Degree Confluence Project (DCP) is a volunteer-based validation dataset that comprises useful information for global land cover map validation. However, there is a problem with using DCP points as validation data for the accuracy assessment of land cover maps. While resolutions of typical global land cover maps are several hundred meters to several kilometers, DCP points can only guarantee an area of several tens of meters that can be confirmed by ground photographs. So, the objective of this study is to create a land cover map validation dataset with added spatial uniformity information using satellite images and DCP points. For this, we devised a new method to semiautomatically guarantee the spatial uniformity of DCP validation data points at any resolution. This method can judge the validation data with guaranteed uniformity with a user’s accuracy of 0.954. Furthermore, we conducted the accuracy assessment for the existing global land cover maps by the DCP validation data with guaranteed spatial uniformity and found that the trends differed by class and region.

Active Collection of Land Cover Sample Data from Geo-Tagged Web Texts

Sample data plays an important role in land cover (LC) map validation. Traditionally, they are collected through field survey or image interpretation, either of which is costly, labor-intensive and time-consuming. In recent years, massive geo-tagged texts are emerging on the web and they contain valuable information for LC map validation. However, this kind of special textual data has seldom been analyzed and used for supporting LC map validation. This paper examines the potential of geo-tagged web texts as a new cost-free sample data source to assist LC map validation and proposes an active data collection approach. The proposed approach uses a customized deep web crawler to search for geo-tagged web texts based on land cover-related keywords and string-based rules matching. A data transformation based on buffer analysis is then performed to convert the collected web texts into LC sample data. Using three provinces and three municipalities directly under the Central Government in China as study areas, geo-tagged web texts were collected to validate artificial surface class of China’s 30-meter global land cover datasets (GlobeLand30-2010). A total of 6283 geo-tagged web texts were collected at a speed of 0.58 texts per second. The collected texts about built-up areas were transformed into sample data. User’s accuracy of 82.2% was achieved, which is close to that derived from formal expert validation. The preliminary results show that geo-tagged web texts are valuable ancillary data for LC map validation and the proposed approach can improve the efficiency of sample data collection.

Active Collection of Land Cover Sample Data from Geo-Tagged Web Texts

Sample data plays an important role in land cover (LC) map validation. Traditionally, they are collected through field survey or image interpretation, either of which is costly, labor-intensive and time-consuming. In recent years, massive geo-tagged texts are emerging on the web and they contain valuable information for LC map validation. However, this kind of special textual data has seldom been analyzed and used for supporting LC map validation. This paper examines the potential of geo-tagged web texts as a new cost-free sample data source to assist LC map validation and proposes an active data collection approach. The proposed approach uses a customized deep web crawler to search for geo-tagged web texts based on land cover-related keywords and string-based rules matching. A data transformation based on buffer analysis is then performed to convert the collected web texts into LC sample data. Using three provinces and three municipalities directly under the Central Government in China as study areas, geo-tagged web texts were collected to validate artificial surface class of China’s 30-meter global land cover datasets (GlobeLand30-2010). A total of 6283 geo-tagged web texts were collected at a speed of 0.58 texts per second. The collected texts about built-up areas were transformed into sample data. User’s accuracy of 82.2% was achieved, which is close to that derived from formal expert validation. The preliminary results show that geo-tagged web texts are valuable ancillary data for LC map validation and the proposed approach can improve the efficiency of sample data collection.