Agricultural Machinery Trajectory Data for Field-Road Classification

Three real agricultural machinery trajectory datasets used to validate a novel classification method. The data was used in a study by Weixin Zhai, achieving an average F1-score improvement of 12.16% over existing methods. The dataset was last updated on May 12, 2026.

Use Cases

• Segmenting agricultural machinery GPS trajectories into field and road segments based on movement patterns.
• Evaluating clustering algorithms for distinguishing weakly connected high-density and low-density trajectory points.
• Analyzing the spatiotemporal evolution patterns of agricultural machinery operations.
• Correcting misclassified trajectory points at field-road intersections using direction statistical features.

Strengths

• Validated on three real agricultural machinery trajectory datasets.
• The proposed method achieved a 20.85% increase in F1-score for field category classification on a wheat harvester dataset.
• Published under the CC-BY-4.0 license, allowing for broad reuse.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The primary file format is JPG, which suggests the data may be presented as images rather than raw trajectory coordinates.

Provenance

Source

figshare

Collection Method

Likely collected from GPS units on agricultural machinery.

Freshness

Last updated 2026-05-12 10:41:16; freshness should be verified.