{"id":120548,"date":"2024-11-11T07:09:08","date_gmt":"2024-11-11T05:09:08","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/reliz-sistemy-raspoznavaniya-teksta-tesseract-5-5-0"},"modified":"2024-11-11T07:09:08","modified_gmt":"2024-11-11T05:09:08","slug":"reliz-sistemy-raspoznavaniya-teksta-tesseract-5-5-0","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/novosti-interneta\/reliz-sistemy-raspoznavaniya-teksta-tesseract-5-5-0","title":{"rendered":"L\u00ebshimi i sistemit t\u00eb njohjes s\u00eb tekstit Tesseract 5.5.0","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"
\u00cbsht\u00eb publikuar versioni i sistemit t\u00eb njohjes optike t\u00eb tekstit Tesseract 5.5.0, q\u00eb mb\u00ebshtet Unicode dhe njohjen e teksteve n\u00eb m\u00eb shum\u00eb se 100 gjuh\u00eb, p\u00ebrfshir\u00eb rusishten, kazakishten, bjellorusishten dhe ukrainishten. Rezultati mund t\u00eb ruhet si tekst i hapur, ashtu edhe n\u00eb formate HTML (hOCR), ALTO (XML), PDF dhe TSV. Fillimisht, sistemi u krijua n\u00eb vitet 1985-1995 n\u00eb laboratorin e kompanis\u00eb Hewlett Packard, n\u00eb vitin 2005 kodi u hap n\u00ebn licenc\u00ebn Apache dhe m\u00eb pas u zhvillua me pjes\u00ebmarrjen e punonj\u00ebsve t\u00eb kompanis\u00eb Google. Tekstet burimore t\u00eb projektit shp\u00ebrndahen n\u00ebn licenc\u00ebn Apache 2.0. <\/p>\n
Tesseract p\u00ebrfshin nj\u00eb utilitar konsol\u00eb dhe bibliotek\u00ebn libtesseract p\u00ebr integrimin e funksioneve t\u00eb njohjes s\u00eb tekstit n\u00eb aplikacione t\u00eb tjera. Nga nd\u00ebrfaqet GUI t\u00eb tret\u00eb q\u00eb mb\u00ebshtesin Tesseract, mund t\u00eb p\u00ebrmendim gImageReader, VietOCR dhe YAGF. Ofrohen dy motor\u00eb njohjeje: klasi, q\u00eb njoh tekstin n\u00eb nivelin e shablloneve t\u00eb simbolit, dhe i ri, i bazuar n\u00eb p\u00ebrdorimin e sistemit t\u00eb m\u00ebsimit t\u00eb makineris\u00eb me rrjete nervore rekurente LSTM, i optimizuar p\u00ebr njohjen e linjave t\u00eb plota dhe q\u00eb lejon nj\u00eb rritje t\u00eb r\u00ebnd\u00ebsishme t\u00eb sakt\u00ebsis\u00eb. Modelet e trajnuara kan\u00eb qen\u00eb t\u00eb publikuara p\u00ebr 123 gjuh\u00eb. P\u00ebr optimizimin e performanc\u00ebs propozohen module q\u00eb p\u00ebrdorin OpenMP dhe instruksione SIMD AVX2, AVX, AVX512F, NEON ose SSE4.1. <\/p>\n
P\u00ebrmir\u00ebsimet kryesore: <\/p>\n
\n
Mb\u00ebshtetja p\u00ebr zgjerimet vektoriale RISC-V V \u00ebsht\u00eb shtuar, mbi t\u00eb cilat jan\u00eb p\u00ebrgatitur optimizimet e assembler-it p\u00ebr sistemet me procesor\u00eb RISC-V.\n
P\u00ebr kur t\u00eb shkruhet rezultati n\u00eb formatin hOCR, \u00ebsht\u00eb siguruar vendosja n\u00eb skedarin q\u00eb krijohet t\u00eb parametrave ocrp_dir dhe ocrp_lang.\n
Kodi \u00ebsht\u00eb p\u00ebrmir\u00ebsuar p\u00ebr t\u00eb formuar skedar\u00ebt n\u00eb formatin hOCR dhe \u00ebsht\u00eb eliminuar ndryshimi i emrave t\u00eb skedar\u00ebve n\u00eb platform\u00ebn Windows.\n
\u00cbsht\u00eb e lejuar t\u00eb jepen vlera simbolike n\u00eb opsionet \u00ab\u2014oem\u00bb dhe \u00ab\u2014psm\u00bb.\n
N\u00eb kod \u00ebsht\u00eb b\u00ebr\u00eb z\u00ebvend\u00ebsimi i funksioneve access dhe _access me metod\u00ebn std::filesystem::exists(). Funksionet tprintf jan\u00eb z\u00ebvend\u00ebsuar me p\u00ebrdorimin e rrjedh\u00ebs tesserr.\n
Mb\u00ebshtetja p\u00ebr platform\u00ebn e m\u00ebsimit t\u00eb makineris\u00eb Tensorflow \u00ebsht\u00eb fshir\u00eb, e cila nj\u00eb her\u00eb ishte implementuar, por nuk ishte p\u00ebrdorur p\u00ebr realizimin e modeleve AI t\u00eb njohjes.\n
Installer-i p\u00ebr platform\u00ebn Windows \u00ebsht\u00eb p\u00ebrmir\u00ebsuar.\n