What Separates Parcel-Sense

Zero-based parcel reconciliation, not a change detection product.

Change detection is valuable. It helps assessment offices identify recent construction, demolition, and site activity. Parcel-Sense is designed for a different question: does the current assessment record agree with what is visible on the ground today?

What change detection is built to do

Change detection generally compares imagery from different dates and identifies areas where something appears to have changed. That is useful for recent additions, recent demolitions, new construction, grading, roof changes, and other events that occur within the available comparison window.

For assessment offices, this can be a strong operational tool. It helps direct staff toward recent activity that may require review. Parcel-Sense does not replace that function.

What Parcel-Sense is built to do

Parcel-Sense performs a zero-based scan of the current parcel inventory. The goal is not to ask what changed between two images. The goal is to ask whether current imagery-derived observations appear consistent with the current assessment record.

This is important because assessment records can drift over time. A structure may have existed for years without being correctly reflected in the record. A building may have been removed before the available change detection window. A parcel may have been coded incorrectly long ago and then carried forward year after year.

Long-standing missing structures may not appear as recent change.
Old demolitions may not trigger a current change detection workflow.
Incorrect vacant land records can persist even when the improvement is visible today.
Legacy record issues can remain hidden unless the inventory is reconciled against current ground conditions.

Where Parcel-Sense fits in the assessment workflow

Parcel-Sense produces a full parcel-level inference dataset. The dataset can be filtered and sorted to surface contradictions between imagery observations and assessment records. For example, a jurisdiction can filter to parcels recorded as vacant land, then review records where the model agreement indicates visible improvements.

The value is not that the system makes final assessment decisions. The value is that it narrows a large parcel inventory into a more workable review path, while leaving final determinations with the jurisdiction.

1. Assessment Record
Existing parcel attributes, improvement records, vacancy indicators, building counts, and related assessment data.
2. Current Imagery
Aerial imagery reviewed at the parcel level using parcel-aware computer vision and spatial logic.
3. Inference Dataset
Model outputs, consensus indicators, footprint square foot estimates, and agreement indicators tied back to parcel records.
4. Staff Review
Filtering and sorting guidance helps staff identify records that warrant additional internal review or field verification.
Side by Side

Two different questions.

Change detection and Parcel-Sense can complement each other because they are designed around different review questions.

Review Question Change Detection Parcel-Sense
What changed recently? Strong fit. May help, but this is not the primary purpose.
What changed between two imagery dates? Strong fit. Not the core workflow.
Does the current record match current ground conditions? Limited, depending on the source of the issue. Primary purpose.
Can long-standing record drift be surfaced? Limited when the issue predates the comparison window. Strong fit.
Does the workflow require historical imagery comparison? Usually yes. No. The scan can begin from current imagery and current assessment records.
Is the service designed as a recurring subscription? Often recurring, depending on vendor and imagery cycle. Primarily a one-time reconciliation event, with future scans available if needed.

A practical scan for records that have drifted over time.

Parcel-Sense is built around a simple assessment problem: records can become inaccurate even when nothing changed recently. A zero-based reconciliation scan helps surface those records without asking the jurisdiction to adopt another long-term platform by default.