Preserve both sides

Pair and preserve the front and back of old photos

Pair front and back scans, draft a local transcription, add confirmed context and export an evidence-preserving archive without uploading family photographs.

Build a memory capsule

Completely free No account or payment Image stays on this device

USE THIS WHENUse this when an old photograph and the writing on its reverse were scanned as separate files and must remain one understandable record.SUPPORTED INPUTUp to 60 front/back scans

What you will getA clear result before technical detail

01Confirmed front-and-back photo records

02A human-reviewed reverse transcription

03Original scans, checksums and a preservation manifest in one ZIP

Private analysis labFiles stay inside this browser

CAPSULE / 60 FILES

Choose front and back photo scansUp to 60 JPEG, PNG or WebP files · name pairs with front/back, recto/verso or F/B

Selection is processed locally

Session diagnosticsNo processing errors

Stored only in this browser tab. Image bytes, filenames and metadata are never included.

No tool error has been recorded in this tab.

No upload or account Originals remain unchangedReview methodology

Useful and careful

What is built into the result.

Important findings come first. Technical fields remain available without taking over the page.

Pairs recto and verso without face recognition

Local OCR creates a draft but never overwrites human judgment

Original files and their SHA-256 identifiers stay intact in the archive

Understand the method

A useful result with its reasoning attached.

This page explains what the lab measures, how to interpret it and where human review remains essential.

Why the back of a photograph belongs with the front

A scanned photograph can lose its most useful context when the reverse side is stored as an unrelated file. Handwritten names, dates, studio marks, album positions and family notes may be more informative than ordinary EXIF metadata. This lab treats both scans as one preservation record rather than two loose images.

AnalyzeImage first looks for conservative filename signals such as front and back, recto and verso, or F and B. Every proposed pairing remains visible. Ambiguous records are marked for review instead of being silently joined.

No face matching
No invented dates or identities
Pairing reason stored in the report
Missing sides remain explicit

A transcription workflow that keeps people in control

Local English OCR can create a starting draft from the reverse scan. Handwriting, faded pencil and curved album pages are difficult for any OCR engine, so the draft is placed in an editable field and labeled with its confidence. Only the text the user leaves in that field is exported as the confirmed transcription.

Caption, date range and place are separate fields. This prevents uncertain handwriting from being mixed with known family context. Approximate dates can remain approximate; the tool never turns a decade or range into a false exact day.

OCR model loaded from AnalyzeImage
Image pixels remain in the browser
Every word can be corrected
Blank OCR results can be entered manually

What the preservation ZIP contains

Each record receives its own folder with the supplied front scan, back scan and a plain-text record. A versioned JSON manifest records the original names, byte sizes, pairing confidence, confirmed notes and SHA-256 file identifiers. These identifiers help detect an accidental file change; they do not establish ownership or authenticity.

The package is designed to remain understandable without AnalyzeImage. Original image formats are retained, the text record opens in ordinary editors and the JSON structure can be imported into a future archive workflow.

Unmodified source scans
Portable text record
Versioned JSON manifest
Local ZIP creation

Three simple steps

Know what happens before you start.

Use this when an old photograph and the writing on its reverse were scanned as separate files and must remain one understandable record.

PairReview suggested front-and-back pairs; ambiguous filenames remain marked for confirmation.
DescribeCorrect the optional local OCR draft and add only context you actually know.
PreserveExport originals, text records, checksums and a versioned manifest in one ZIP.

RESULT ORDER

01 · Confirmed front-and-back photo records02 · A human-reviewed reverse transcription03 · Original scans, checksums and a preservation manifest in one ZIP04 · Limits and next step

Clear before you rely on it

Questions this tool should answer upfront.

Short answers keep important privacy, evidence and professional-use limits visible.

01Does the tool recognize people in old photographs?

No. It does not perform face recognition or infer a person's identity. You add only the names and context you already know.

02Are family photographs uploaded for OCR?

No. The OCR worker, language data and image processing run in the browser. The selected scan is not sent to an AnalyzeImage upload endpoint.

03Can I keep an uncertain date such as the 1950s?

Yes. Date and caption fields are plain text so uncertainty can be preserved instead of converted into a misleading exact date.

Method referencesLibrary of Congress personal digital archiving resources IPTC Photo Metadata Standard

Continue when useful

Your next step, without starting over.

Move to another page only when its outcome matches what you need.

Inspect embedded metadataReview dates, captions and camera fields separately.Understand IPTC and XMPChoose portable descriptive fields carefully.

Important limitations

Filenames can suggest a pair but cannot prove it · Handwriting OCR can be inaccurate and must be reviewed · The package preserves supplied context; it cannot identify unknown people, places or dates

Full limitations