0004.jpg
new
→
triaged
→
uploaded
→
enriched
→
extracted
→
▶ closed
→
remove
ARK
Name
2DbcsqZjYYgfdN9NvZZexW
https://n2t.net/ark:/11635/2DbcsqZjYYgfdN9NvZZexW
https://scanstation.ai/ark/11635/2DbcsqZjYYgfdN9NvZZexW
File
| Order | 4 |
|---|---|
| Dimensions | 7968 × 5984 |
| Orientation | 0 |
| S3 Key | marac/003/derivatives/0004.jpg |
| Source Path | marac/003/derivatives/0004.jpg |
AI Enrichment
Summary
Back of a used divided-back U.S. postcard with dense handwritten message, green Washington postage stamp, and circular postmark likely from Akron, Ohio, addressed to a Mrs. at 2651 Workman St., Los Angeles, California; postmark appears to date to 1932.
Places
Los Angeles, California
Akron, Ohio
Type
people_photo
document
handwritten_note
photo
text_page
postcard
Tags
No tags assigned.
Primary: postcard
No description yet.
Subject: image:01KQHRN0FYZPRDXKYWS3F6HXHG
18 predicates
| Predicate | Best value | Conf. | Source |
|---|---|---|---|
| ai:denseSummary |
Back of a used divided-back U.S. postcard with dense handwritten message, green Washington postage stamp, and circular postmark likely from Akron, Ohio, addressed to a Mrs. at 2651 Workman St., Los Angeles, California; postmark appears to date to 1932.
|
1% | enrich_from_thumbnail@1.0 |
| ai:documentLikely |
yes
ai-tools heuristic on scan.storage:marac/003/0004.jpg
|
1% | ai-tools@1.0 |
| ai:handwrittenText | yes | 1% | enrich_from_thumbnail@1.0 |
| ai:hasText | yes | 1% | enrich_from_thumbnail@1.0 |
| ai:highResGoal |
read_handwriting
Loss of sender/recipient details, date/place of mailing, and message content that may contain names, locations, and contextual cataloging evidence.
|
1% | enrich_from_thumbnail@1.0 |
| ai:highResTarget |
entire handwritten message in left correspondence panel
High-resolution target from thumbnail enrich
|
1% | enrich_from_thumbnail@1.0 |
| ai:isFilledForm | yes | 1% | enrich_from_thumbnail@1.0 |
| ai:isForm | yes | 1% | enrich_from_thumbnail@1.0 |
| ai:ocrText |
Tesseract via ai-tools (11 words, mean conf 33.8)
|
0% | ai-tools@1.0 |
| ai:peopleLikely |
yes
ai-tools heuristic on scan.storage:marac/003/0004.jpg
|
1% | ai-tools@1.0 |
| ai:photoLikely |
yes
ai-tools heuristic on scan.storage:marac/003/0004.jpg
|
1% | ai-tools@1.0 |
| ai:photoWithTextLikely | |||
| ai:pixelsDone | |||
| ai:speculation |
This is a mailed U.S. divided-back postcard sent from Akron, Ohio to a woman in Los Angeles, California.
Visible divided postcard back, U.S. stamp, apparent Akron postmark, and address reading Los Angeles, California.
|
1% | enrich_from_thumbnail@1.0 |
| ai:typedText | yes | 1% | enrich_from_thumbnail@1.0 |
| dcterms:spatial |
Los Angeles, California
Akron, Ohio
|
1% | aggregated |
| dcterms:type |
people_photo
found 1 face(s)
|
5% | ai-tools@1.0 |
| foaf:Person |
Mrs Pull-hingrin
Recipient line appears to begin 'Mrs' followed by an unclear surname; handwriting is difficult to decipher at thumbnail scale.
|
1% | enrich_from_thumbnail@1.0 |
Subject: image:01KQHRN0FYZPRDXKYWS3F6HXHG
0 claims
No claims recorded yet for this subject.
closed
— No more image calls. Item workflow consumes image claims/text/metadata for catalog-level analysis.
| Transition | Description | |
|---|---|---|
| upload | Pushes realPath (the ai-tools crop) to S3 as the canonical image. When defaults.full_frame_key is present, also archives the original mat under originals/ on S3 for re-processing / QR re-extraction during Python dev. The full-frame archive is planned to drop for real-client tenants. | |
| external_upload | Sets defaults["storage"]["mode"] = "external_url". Used by IIIF imports so the image remains accessible via its external URL and subsequent transitions can run against it directly. Skips triage — the UPLOADED place lets enrich run against the remote URL. | |
| triage | When fired from NEW: local-file mode. ai-tools reads defaults.full_frame_key + defaults.sidecar_key from the shared dir, produces the canonical crop + thumbnail in derivatives/, and returns deskew/type/detections. realPath is repointed at the crop. When re-run from TRIAGED: same, idempotent. Non-URL mode is the only path browser captures take. Failures keep the image in NEW (async retry). | |
| enrich | Single cheap AI call (~$0.0004) against a ≤512px thumbnail, with triage OCR + flags in context. Populates title/description/subjects/people/places/dense_summary and text-type flags. Can run from TRIAGED (using ssai image_view via imgProxy, skipping S3 upload) or from UPLOADED. Re-runnable from enriched for debugging. | |
| extract | The optional second image call. Runs only when enrich_from_thumbnail sets pixelsDone=false. Dispatches by extraction goal/signals: filled forms → ExtractMetadata; handwriting → TranscribeHandwriting; dense print/layout → OcrMistral. Re-runnable. | |
| close | Bookkeeping only. Requires pixelsDone=true. The item workflow performs text-only catalog analysis across all images. | Apply |
Subject: image:01KQHRN0FYZPRDXKYWS3F6HXHG
29 claims
enrich_from_thumbnail@1.0
19
| Predicate | Value | Conf. | Basis |
|---|---|---|---|
| ai:denseSummary |
Back of a used divided-back U.S. postcard with dense handwritten message, green Washington postage stamp, and circular postmark likely from Akron, Ohio, addressed to a Mrs. at 2651 Workman St., Los Angeles, California; postmark appears to date to 1932.
|
1% | — |
| ai:handwrittenText | yes | 1% | — |
| ai:hasText | yes | 1% | — |
| ai:highResGoal |
read_handwriting
|
1% |
Loss of sender/recipient details, date/place of mailing, and message content that may contain names, locations, and contextual cataloging evidence.
|
| ai:highResTarget |
entire handwritten message in left correspondence panel
|
1% |
High-resolution target from thumbnail enrich
|
| ai:highResTarget |
recipient name and street address in right address panel
|
1% |
High-resolution target from thumbnail enrich
|
| ai:highResTarget |
postmark text and date
|
1% |
High-resolution target from thumbnail enrich
|
| ai:highResTarget |
publisher or serial number area near lower right
|
1% |
High-resolution target from thumbnail enrich
|
| ai:highResTarget |
any text beneath or around the stamp
|
1% |
High-resolution target from thumbnail enrich
|
| ai:isFilledForm | yes | 1% | — |
| ai:isForm | yes | 1% | — |
| ai:pixelsDone |
Thumbnail shows substantial handwritten message, address, stamp, and postmark, but much text is too small or obscured to transcribe reliably.
|
||
| ai:speculation |
This is a mailed U.S. divided-back postcard sent from Akron, Ohio to a woman in Los Angeles, California.
|
1% |
Visible divided postcard back, U.S. stamp, apparent Akron postmark, and address reading Los Angeles, California.
|
| ai:speculation |
The card likely dates to the early 1930s.
|
1% |
Postmark appears to show year '1932', though the small numerals are not fully secure at thumbnail scale.
|
| ai:typedText | yes | 1% | — |
| dcterms:spatial | Los Angeles, California | 1% |
Address panel clearly includes 'Los Angeles' and 'California'.
|
| dcterms:spatial | Akron, Ohio | 1% |
Postmark appears to read 'AKRON O.'; legibility is limited by thumbnail resolution.
|
| dcterms:type |
postcard
|
1% | — |
| foaf:Person |
Mrs Pull-hingrin
|
1% |
Recipient line appears to begin 'Mrs' followed by an unclear surname; handwriting is difficult to decipher at thumbnail scale.
|
ai-tools@1.0
10
| Predicate | Value | Conf. | Basis |
|---|---|---|---|
| ai:documentLikely | yes | 1% |
ai-tools heuristic on scan.storage:marac/003/0004.jpg
|
| ai:hasText |
Tesseract via ai-tools (11 words, mean conf 33.8)
|
||
| ai:peopleLikely | yes | 1% |
ai-tools heuristic on scan.storage:marac/003/0004.jpg
|
| ai:photoLikely | yes | 1% |
ai-tools heuristic on scan.storage:marac/003/0004.jpg
|
| ai:photoWithTextLikely |
ai-tools heuristic on scan.storage:marac/003/0004.jpg
|
||
| dcterms:type |
people_photo
|
5% |
found 1 face(s)
|
| dcterms:type |
document
|
4% |
document outline covers 83% of frame; visual text regions suggest document-like content; handwritten text still looks document-like
|
| dcterms:type |
handwritten_note
|
4% |
OCR found text with low confidence, which often means handwriting
|
| dcterms:type |
photo
|
2% |
face detection usually implies a photograph
|
| dcterms:type |
text_page
|
1% |
visual text regions suggest text blocks
|