Encoding and compressing hexagonal raster data by the Gosper curve. Issue 6 (9th July 2019)

• Record Type: Journal Article
• Authors: Xin, Rui; Ai, Tinghua; Ai, Bo
• Is Part Of: Transactions in GIS. Volume 23:Issue 6(2019)
• Page Start: 1204
• Page End: 1231
• Publication Date: 2019-07-09
• Subjects: Geographic information systems -- Periodicals; 910.285
• DOI: 10.1111/tgis.12569
• Languages: English
• ISSNs: 1361-1682

Abstract

Abstract: This article mainly introduces a class of encoding and compression methods for hexagonal raster data. A new encoding mode is established with the introduction of the Gosper curve, which has good spatial aggregation. On this basis, straightforward encoding, lossless coding compression, and lossy coding compression can be carried out. First, the bidirectional correlation between the Gosper curve and the hexagonal grid data is established, and the Gosper coding value of each raster unit is determined. Then, the lossless compression of the data is completed by performing run‐length encoding on the coding set. Following the Gosper curve trend, partial grid units are semantically changed via adjacent fusion under certain threshold constraints. The run‐length encoding is carried out again to complete the lossy compression and further reduce the amount of encoding required. A break‐detection step is added during the process to avoid regional breaks. The experimental results show that the proposed method can efficiently compress data. Additionally, data compression of multi‐resolution and varying fusion thresholds is explored to summarize general patterns. Finally, the method proposed is compared with traditional row‐coding compression in hexagonal and quadrilateral grids to highlight its advantages for hexagonal raster data.