Accuracy Evaluation Method for Vector Data Based on Hexagonal Discrete Global Grid

Yue Ma, Guoqing Li, Long Zhao, Xiaochuang YaoISPRS Int. J. Geo Inf 2024 · OpenReview

TL;DR

This paper introduces a novel accuracy evaluation system for vector data that has been converted into a hexagonal Discrete Global Grid System (DGGS). This system addresses the challenges of data redundancy, geometric distortion, and discontinuity that arise when representing vector data on a grid, offering a standardized method to assess the quality of this transformation.

What Problem Does This Paper Address?

The ever-increasing volume of geographic spatial data necessitates efficient storage, organization, and management. Discrete Global Grid Systems (DGGS) offer a standardized way to handle this, with hexagonal grids being particularly promising for representing vector data due to their equal-area properties and ability to mitigate issues like data redundancy and geometric deformation common in traditional gridded formats. However, a significant gap exists in the field: there's no comprehensive, quantifiable system to evaluate the accuracy of vector data after it's been converted into these hexagonal DGGS. This paper aims to fill that void by proposing a systematic approach to assess the accuracy of gridded vector data, covering geographical deviation, geometric features, and topological relationships.

Methods

The authors propose a multi-faceted accuracy evaluation system for hexagonal-grid-based vector data. The core of their approach involves:

  • Gridding Methods: They outline different gridding strategies tailored for various vector data types (points, lines, polygons). For points, a "drop point" method is used. For lines, the Bresenham algorithm is adapted for hexagonal grids. Polygons are handled using a combination of Bresenham for edges and a scanline fill algorithm for the interior.
  • Coordinate System: A unique IJ coordinate system is established for the hexagonal grid, facilitating conversion between geographic and grid coordinates.
  • Accuracy Evaluation Metrics: The system is built upon three key aspects:
  • Geographical Deviation: This measures the Euclidean distance between original vector points and the centroids of their corresponding grid cells.
  • Geometric Features: For lines, this includes angular distortion and length distortion. For polygons, it encompasses area ratio, overlap area ratio, perimeter ratio, minimum external circle area ratio, and minimum external rectangle area ratio.
  • Topological Distortion: This assesses how the spatial relationships between vector features are preserved or altered during the gridding process. For points, it focuses on how features that were originally disjoint might become "equal" (fall into the same grid cell). For lines, it checks for distortions like "disjoint becomes overlap" or "open line becomes closed line." For polygons, it examines cases like faces disappearing, splitting, becoming adjacent, or losing enclosing relationships.
  • Hierarchical Evaluation: The system leverages the multi-resolution nature of DGGS, allowing for evaluation at different grid levels and suggesting that increasing the grid level can resolve many distortions.

Datasets

The paper utilizes real-world datasets for evaluation:

  • Point Data: Subway coordinate points in Beijing, China.
  • Line Data: G45 and Daguang Expressway data in Beijing, China.
  • Polygon Data: Administrative boundary data of Beijing, China.

These datasets are used to demonstrate the application of the proposed accuracy evaluation system and to showcase the effectiveness of gridding at different levels.

Key Insights and Ideas

  • Hexagonal Grids as a Solution: The paper strongly advocates for hexagonal grids within DGGS as a superior method for representing vector data, effectively addressing long-standing issues of data redundancy and geometric distortion inherent in traditional gridded approaches.
  • Comprehensive Accuracy Framework: The proposed three-tiered evaluation system (geographical, geometric, topological) provides a holistic view of data quality after gridding, moving beyond simple positional accuracy.
  • Topological Preservation is Crucial: The emphasis on topological distortion highlights that simply preserving coordinates isn't enough; the relationships between features are vital for meaningful spatial analysis.
  • Hierarchy as a Tool for Correction: The multi-resolution nature of DGGS is not just for storage but is actively used as a mechanism to resolve topological distortions by increasing the grid level.
  • Data Scale Matters: The paper implicitly shows how the scale of the original vector data and the chosen grid resolution directly impact the accuracy metrics.

Key Results

The paper presents experimental results demonstrating the effectiveness of its proposed accuracy evaluation system:

  • Point Data: For Beijing subway data, increasing the grid level from 14 to 15 resolved most topological distortions. The average coordinate difference before and after conversion was 0.096° (8.624 km).
  • Line Data: For highway data, increasing the grid level consistently improved accuracy across all metrics (Geographical Deviation, Angular Distortion, Length Distortion). For instance, Geographical Deviation decreased from 12.67 km at level 10 to 0.76 km at level 13.
  • Polygon Data: For Beijing administrative boundaries, accuracy also improved with higher grid levels. The Composite Index for geometric accuracy increased from 0.9647 at level 10 to 1.0135 at level 12.

Limitations and Future Work

The authors acknowledge several limitations and suggest future directions:

  • Artificial Index Weights: The determination of weights for evaluation indicators (e.g., in the Analytic Hierarchy Process for polygon shape similarity) is currently somewhat artificial and may require modification based on specific application scenarios.
  • Generalizability: While the system is presented as generalizable, further research is needed to refine feature indexes and similarity calculation methods for specific applications to improve accuracy.
  • Further Development: Future work should focus on developing more precise and accurate evaluation index systems, exploring new applications for hexagonal-grid-based data, and investigating integration with other data management and storage systems. The paper also suggests the need for more research into improving the accuracy of gridding itself.

Summary updated

Abstract

With the continuous advancement of technology for obtaining geographic spatial data, the accumulated volume of such data has been increasing, thus imposing higher demands on the storage, organization, and management of such data. As a new form of data management, the Discrete Global Grid System (DGGS) provides standardized descriptions and the exchange of geographic information on a global scale, enabling the efficient storage and application of large-scale global spatial data. Constituting a traditional type of GIS spatial data, vector data have advantages such as clear positions, implicit attributes, and suitability for map output. The representation of vector data in the global discrete grid network based on an equal-area projection, such as the hexagonal grid, fundamentally solves problems such as data redundancy, geometric deformation, and data discontinuity that arise when representing multiple vector data in a gridded format. This paper proposes different gridding methods for various types of vector data, and a quantifiable accuracy evaluation system is established from the perspectives of geographical deviation, geometric features, and topological relationships, to evaluate the accuracy of the gridded vector data, covering all types of gridded vector data based on the hexagonal grid. The evaluation method is generally applicable to all hexagonal-grid-based gridded vector data, and can be generalized based on application scenarios, for evaluating the usability of hexagonal grid vector data.
BibTeX
@article{ma2024accuracy,
  author = {Yue Ma and Guoqing Li and Long Zhao and Xiaochuang Yao},
  title = {{Accuracy Evaluation Method for Vector Data Based on Hexagonal Discrete Global Grid}},
  journal = {ISPRS Int. J. Geo Inf.},
  year = {2024},
  url = {https://openreview.net/forum?id=eiqVZ6Gtnz},
}

Loading related content...

Figure