Positional Accuracy Assessment of ALOS Satellite Orthoimagery Using Spatial Statistics

A methodology paper incorporates spatial statistics techniques like Nearest Neighbor and Ripley's K function for cartographic quality control. The work evaluates an orthoimage generated from a PRISM scene of the ALOS satellite. Results show the orthoimage is accurate at a 1:25,000 scale and classified as Class A according to the Brazilian positional accuracy standard.

Use Cases

• Evaluating spatial distribution patterns of point data based on the described Nearest Neighbor and Ripley's K function methods.
• Analyzing bias in positional discrepancies using average directional vectors and circular variance as mentioned in the description.
• Implementing a positional quality control workflow based on the proposed methodology, which includes sampling planning and normality testing.
• Classifying the positional accuracy of spatial data according to established standards like the Brazilian standard referenced in the paper.

Strengths

• The methodology is validated with a practical experiment using an orthoimage from the ALOS satellite's PRISM sensor.
• Results provide a concrete accuracy classification (Class A) at a defined map scale of 1:25,000.
• The work proposes a complete methodology integrating sampling, spatial pattern analysis, bias testing, and accuracy classification.

Limitations

• The actual dataset (e.g., the orthoimage or discrepancy samples) is not directly available; the listing describes a methodology paper.
• Column-level documentation and sample data are unavailable, making direct data inspection impossible.
• Row count, file formats, and data size are unknown, limiting suitability assessment for direct ML/AI application.

Provenance

Source

Afonso de Paula dos Santos via paperswithcode.

Collection Method

Methodology proposed and tested using an orthoimage from the ALOS satellite.

Geography

The example uses the Brazilian positional accuracy standard, but the specific geographic coverage of the evaluated orthoimage is not stated.