Inspection Image Matcher

Matches drone and field inspection images to known defects.

Industry	Level	Difficulty	Recommended mode
Energy & utilities	3 of 5	Advanced	hnsw

Coming soon

A complete, runnable Inspection Image Matcher reference implementation is in progress. The architecture and core code below already work with SrvDB 0.3.

The problem

• Drone inspections produce huge image sets.
• Defects are subtle.
• Review is slow.

How SrvDB powers Inspection Image Matcher

1. Images embedded with defect labels.
2. New images retrieve similar defects.
3. hnsw for scale; sq8 for size.
4. Reviewed by engineers.

What changes

Today	With Inspection Image Matcher on SrvDB
Manual image review	Prioritised images
Subtle defects missed	Similar defects surfaced
Huge archives	Compact vectors

How it works

1. Embed labelled images.
2. Embed new images.
3. Match.
4. Review.
5. Add labels.

SrvDB features it uses

• HNSW graph (hnsw): approximate search that visits a small part of large collections; tune with set_ef_search().
• SQ8 compression (sq8): stores each value in one byte, 4× smaller than full precision.
• search_batch: runs many queries at once, in parallel inside Rust.
• Crash-safe persist(): after it returns, data survives a crash or power loss and reopens exactly as committed.

Use cases

• Insulator damage
• Corrosion
• Vegetation encroachment
• Solar panel defects
• Wind blade damage
• Pipeline inspection

Core code

embed() is your embedding model (for example sentence-transformers, or a function that turns your records into feature vectors).

import json
import srvdb

db = srvdb.Database("./inspection-image-matcher-db", dimension=512, mode="hnsw")

def ingest(items):
    """items: list of (id, text, metadata dict)"""
    db.upsert(
        ids=[i for i, _, _ in items],
        embeddings=[embed(t) for _, t, _ in items],
        metadatas=[json.dumps(m) for _, _, m in items],
    )
    db.persist()                      # crash-safe checkpoint

def similar_defects(image_description, k=5):
    hits = db.search(embed(image_description), k)
    return [(json.loads(db.get(i)), round(score, 3)) for i, score in hits]

Limits and honest notes

• Needs a validated image embedding model.

Build roadmap

1. Prototype: load a sample of real data with flat mode and check that searches return what an expert expects.
2. Measure: if you move to an approximate mode, compare its results with flat on the same queries (recall).
3. Pipeline: add continuous upsert and decide when to call persist().
4. Harden: test restarts and crashes, add logging, review every output path with domain experts.
5. Deploy: ship the database directory with the application or service.

All Energy & utilities projects

Level	Project	What it does
1 · Beginner	Procedure Finder	Operators find the right procedure from a description.
2 · Intermediate	Outage Analyzer	Finds similar past outages and their causes.
3 · Advanced	Inspection Image Matcher (this page)	Matches drone and field inspection images to known defects.
4 · Expert	Field Operations Agent	An offline agent for crews with procedures, asset history and notes.
5 · Frontier	Grid Analog Forecaster	Finds historical analogues of current weather and load patterns.

See also: Use cases · Live examples (coming soon)