Las Flores Water Company · Altadena, California · 2025

FieldSight

Automated Post-Disaster Field Image Classification System

An ML-powered tool that integrates directly into field data collection workflows — enabling crews to photograph a site and receive an automatic classification of burned, missing, debris, or intact condition, embedded in the record.

Scroll
21
Survey Points Evaluated
Pilot deployment
120
Field Images Processed
via Survey123
2
PyTorch Models
Meter + site condition
~1K
Points for Full Dataset
Awaiting inference
Overview

Turning field photos into decisions

FieldSight emerged from the 2025 Eaton/Altadena wildfire response, when Las Flores Water Company needed to assess hundreds of water infrastructure sites across a fire-damaged landscape — quickly, consistently, and with limited personnel.

The system works inside the data collection tools field crews already use. A technician photographs a site — the meter box, the surrounding lot — and FieldSight automatically classifies what it sees, writing the result directly into the survey record without interrupting the workflow.

Rather than requiring a data analyst to manually review hundreds of images after the fact, every photo becomes a structured data point the moment it is taken.

Manual Review Bottleneck

Hundreds of photos reviewed by hand — slow and inconsistent under crisis conditions

Unstructured Photo Data

Raw images in Survey123 carry no condition data unless manually labeled

Priority Ranking Needed

Crews need an ordered dispatch list — not a photo archive to interpret

Limited Training Labels

Only ~20 manually labeled examples available to train the classifiers

Alicia Ave Tonia Ave Laurice Destroyed (10) Damaged (11) Altadena Survey Area 21 evaluated sites · Alicia Ave corridor

Classification Categories

Four conditions, one photo

FieldSight classifies every field image into one of four structural condition categories — enabling automated population of data collection records without manual review.

Burned

Fire Damage Confirmed

Structural or meter components show direct fire damage — charring, melting, or combustion evident in the image. Highest replacement priority.

10
Missing

Component Absent

Expected infrastructure element is not present — meter, connection, or equipment removed, displaced, or destroyed beyond visual recognition.

Debris

Site Obstructed

Infrastructure may be present but is obstructed by rubble, fallen material, or post-fire remnants. Crew access and condition assessment pending clearance.

9
Not Burned

No Fire Damage

Infrastructure appears undamaged. Site has been cleared of debris. Standard service restoration procedures apply — lower dispatch priority.

12
Methodology

A six-stage pipeline

From GPS-tagged field photograph to ranked priority list — FieldSight processes images through a reproducible automated workflow built on PyTorch and ArcGIS Pro.

01

Field Collection

GPS-tagged photos captured via Survey123 at approximately 60 survey points

02

Data Preparation

ArcGIS Pro processes spatial data; photos exported and organized by site

03

Manual Labeling

Approximately 20 survey points receive expert ground-truth labels for training

04

Model Training

Two binary PyTorch classifiers trained: meter condition and site condition

05

Priority Scoring

Confidence metrics combined into a PriorityScore; uncertain predictions flagged for review

06

Dashboard

Predictions joined to ArcGIS feature class and displayed in Streamlit

Meter Condition Model

Binary PyTorch image classifier

Evaluates each meter photograph to classify whether the water meter unit survived or was destroyed — the primary driver of infrastructure replacement priority.

Destroyed Class 0 10
Damaged Class 1 11

Site Condition Model

Binary PyTorch image classifier

Assesses the surrounding property — whether the site remains in ruins with debris present or has been cleared — informing crew safety and access logistics.

Ruins / Debris Class 0 9
Cleared Class 1 12
Results

Priority rankings

All 21 evaluated sites ranked by combined confidence score. Sites with a destroyed meter receive the highest priority for crew dispatch.

Address Meter Site Score
Meter Condition Distribution
21 sites
Destroyed 10
Damaged 11
Site Condition Distribution
21 sites
Ruins / Debris 9
Cleared 12
Priority Distribution
Critical — Score 60 6
High — Score 50 4
Medium — Score 45 3
Standard — Score 35 8
Spatial View

Sites mapped by urgency

Each marker represents a surveyed address in Altadena, CA. Color encodes priority score — click any marker to inspect site-level predictions.

Review Queue

Sites flagged for human review

Five sites received low-confidence predictions from one or both models. These are automatically flagged for manual inspection before dispatch decisions are made.

Object ID Meter Prediction Meter Conf. Site Prediction Site Conf. Priority Score Status
#619 destroyed 83.9% cleared 57.6% low 57 Needs Review
#1071 damaged 85.7% ruins 61.4% low 52 Needs Review
#718 damaged 70.9% low ruins 59.7% low 52 Needs Review
#596 damaged 63.1% low cleared 99.7% 42 Needs Review
#964 damaged 75.9% borderline cleared 63.4% low 42 Needs Review
Complete Dataset

All 21 evaluated survey points

Full inference output from the pilot deployment — including APN, service connection length, confidence scores, and priority ranking for every evaluated site.

# Address APN Conn. Length (ft) Meter Meter Conf. Site Site Conf. Review Priority