Ci bi mu melGis bi ci topp
Ni ñuy xamee këyitu jaxasoo ci làkk wu yomb ak IA
Aplikaasioŋ yi
GUIDE ci aplikaasioŋ yi
AI supports document recovery through image enhancement, computational imaging, text recognition, and candidate restoration.
Déglul tegtal biiIA-nettali · ~2 min
These methods can reveal or organize evidence, but model predictions must be checked against the source and expert scholarship.
The Vesuvius Challenge uses X-ray CT scans and machine learning to read carbonized Herculaneum scrolls without physically unrolling them. The pipeline reconstructs scroll geometry, segments papyrus layers, flattens their surfaces, and detects ink-related texture; carbon ink and papyrus can be difficult to distinguish because their densities are similar. Ithaca is a deep neural network for restoring missing text and estimating the geographic and chronological attribution of ancient Greek inscriptions; its outputs are candidate interpretations for historians. Handwritten text recognition (HTR) transcribes handwriting into searchable text, while image-enhancement systems address stains, fading, and other degradations. Multispectral imaging captures a document under multiple wavelength bands, including infrared; comparing bands can expose ink or degradation not visible in ordinary color images. A study of historical document restoration used infrared information to isolate degradation and reconstruct a cleaned color image; it does not establish AI reading of erased palimpsests. Each recovery is evidence-informed and should retain access to the original scan.
Vesuvius Challenge dafa jëfandikoo jàngu masin ngir jàng téere Herculaneum yu lakk yi ci CT scanner yi te duñu leen dindi
Ithaca bu DeepMind defaraat mbind yi réeroon ci mbindu Grees yu yàgg yi yàqu, ba noppi xayma seen bis
Archive yi dañuy jëfandikoo xàmmee mbind yiñ bind ak loxo ngir mëna bind bataaxal yu yàgg yi ci base de done yuñ mëna seetee
Nataalu spectre yu bari ak juntuwaay yu bees yi wonenañu ay mbind yuñ efaase ci ay palimpsest fu ñu xoosu pargamin bi ba noppi jëfandikoowaat ko
Improved imaging and recognition can make fragile collections more accessible, but preservation requires storing original scans, recording processing steps, and labeling model-generated readings as hypotheses. Human review remains essential where a mistaken restoration could change historical interpretation.
Bulletin bu amul fayda
Ñatti jaar-jaari IA yuñ firndeel bis bu nekk ci ayu-bis bi, ñu bind leen ci Àngle bu leer. Gratis ba fàww, amul yëgle.
Benn email bis bu nekk ci ayu-bis bi. Benn klik nga dindi abonemaa bi. Dunu musa jaay wala séddoo sa adres.
Nattal sa bopp
Feedback ci saasi ci bépp tontu, ak sertifikaa buñ mëna séddoo ak ID buñ mëna saytu soo jàllee ci njàng mi.
Jàppale njàngum IA te doo fay. AI Understanding mbootaay la buy def te Yàlla tax — amul yëgle, amul paywall, musul am. Defal ab sarax
AI supports document recovery through image enhancement, computational imaging, text recognition, and candidate restoration. These methods can reveal or organize evidence, but model predictions must be checked against the source and expert scholarship.
Scroll yi dañu yomba dagg, moo tax scanner CT ak rayon X ak gis-gis encre IA taxna gëstukat yi mëna leen jàng ci net bi.
Ginaaw encre carbone bi ak papyrus bi daanaka dañuy nuru benn densité ci rayon X yi, IA dafa wara gis wuute yu ndaw yi ci texture bi, moo gën ñu gis wuute bu leer.
Ithaca dafa digal ñu defaraat mbindu Grek yi yàqu, ba noppi wax bis yi ak barab yi ñu ko jëlee.
Sistem HTR yi dañuy jàng bind yu yàgg yi ba noppi soppi leen def ko mbindu digital, suko defee ñu mëna seetee ci archive yu bari.
Ñu ngi leen tàggat ci corpus yu yaatu, xeetu nit ñi dañuy jëfandikoo contexte ngir rang kandidaa yiñ wara defaraat ci passage yu yàqu yi.
Weyal di jàng
Tann nañu yeneen njiit ngir topic bii
Ci bi mu melGis bi ci topp
Ni ñuy xamee këyitu jaxasoo ci làkk wu yomb ak IA
Aplikaasioŋ yi