[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"ssr-repo-dspace000002403":3},{"id":4,"title":5,"authors":6,"files":8,"advisors":53,"authors_detail":54,"advisors_detail":58,"publisher":59,"publish_place":60,"language":61,"type":62,"type_slug":63,"year":64,"department":65,"study_program":66,"keywords":67,"keywords_detail":71,"abstract":78,"abstract_html":79,"url":80},"repo-dspace000002403","Implementation Of Tesseract Ocr In Pesantren Books Using Image Processing And Spell Correction At Ppss Nurul Huda",[7],"Abdul Rohman",[9,14,17,20,23,26,29,32,35,38,41,44,47,50],{"chapter":10,"requires_login":11,"label":12,"available":13},"abstrak",false,"Abstrak",true,{"chapter":15,"requires_login":11,"label":16,"available":11},"pendahuluan","Pendahuluan",{"chapter":18,"requires_login":11,"label":19,"available":13},"bab1","Bab I",{"chapter":21,"requires_login":11,"label":22,"available":13},"bab2","Bab II",{"chapter":24,"requires_login":13,"label":25,"available":13},"bab3","Bab III",{"chapter":27,"requires_login":13,"label":28,"available":13},"bab4","Bab IV",{"chapter":30,"requires_login":13,"label":31,"available":13},"bab5","Bab V",{"chapter":33,"requires_login":13,"label":34,"available":11},"bab6","Bab VI",{"chapter":36,"requires_login":13,"label":37,"available":11},"bab7","Bab VII",{"chapter":39,"requires_login":11,"label":40,"available":11},"daftar_pustaka","Daftar Pustaka",{"chapter":42,"requires_login":13,"label":43,"available":11},"lampiran","Lampiran",{"chapter":45,"requires_login":13,"label":46,"available":11},"jurnal","Jurnal",{"chapter":48,"requires_login":11,"label":49,"available":13},"open","Akses Terbuka",{"chapter":51,"requires_login":13,"label":52,"available":11},"full","Dokumen Lengkap",[],[55],{"name":7,"nim":56,"slug":57},"1941720022","abdul-rohman",[],"Politeknik Negeri Malang","Malang","Inggris","Skripsi (D4)","skripsi-d4",2023,"D-IV TEKNIK INFORMATIKA",null,[68,69,70],"OCR (Optical Character Recognition)","Tesseract","Spell Correction",[72,74,76],{"name":68,"slug":73},"ocr-optical-character-recognition",{"name":69,"slug":75},"tesseract",{"name":70,"slug":77},"spell-correction","The research tackles the issue of digitizing Pesantren books, which frequently contain complex Arabic content, to preserve traditional Islamic education and make it accessible in the digital age. In terms of technical implementation, the goal is to implement Tesseract OCR for Pesantren books and ensure that it can accurately recognize Arabic on Pesantren books, as well as the impact of preprocessing and post-processing. The second is to determine how pre-processing and post-processing techniques affect OCR accuracy. For The third goal is to assess and recommend improvements to the recognition accuracy of Tesseract OCR for Pesantren books. These goals and objectives guide this research as a whole, providing a structured approach to addressing the challenges and opportunities of digitizing Pesantren literature. The research entails gathering a diverse set of Pesantren books that represent various styles, fonts, and layouts commonly found for data collection. This dataset serves as the foundation for experimentation. The technical implementation entails incorporating the Tesseract OCR into the system and configuring it to recognize Arabic scripts, ensuring compatibility with the diverse content of Pesantren books. They are also applied to images of Pesantren book pages for pre-processing techniques such as binarization and noise removal. In addition, post-processing includes the implementation of a spell correction module. This module corrects OCR errors and improves the accuracy and readability of the extracted text. The core of the research is evaluating OCR accuracy. The study uses text similarity and WER to quantify the impact of pre-and post-processing techniques on OCR results. The data collected during the evaluation phase is analyzed to conclude the performance of system in digitizing Pesantren books. This analysis serves as the foundation for the research findings and recommendations. As a result, the research successfully implemented Tesseract OCR for Pesantren books. This entailed configuring Tesseract to recognize Arabic as a critical achievement for handling the mixed-script content commonly found in pesantren books. According to the findings, pre-processing techniques significantly improved OCR accuracy. The evaluation of OCR accuracy revealed significant improvements when pre-processing techniques were used. According to the findings, the most effective approach involved a combination of pre-processing and Tesseract OCR, which resulted in the highest text similarity and the lowest word error rate. Overall, the findings emphasize the importance of pre-and postprocessing techniques in improving OCR accuracy for Pesantren books, as well as practical approaches for digitizing this valuable cultural and educational content.","\u003Cp>The research tackles the issue of digitizing Pesantren books, which frequently contain complex Arabic content, to preserve traditional Islamic education and make it accessible in the digital age. In terms of technical implementation, the goal is to implement Tesseract OCR for Pesantren books and ensure that it can accurately recognize Arabic on Pesantren books, as well as the impact of preprocessing and post-processing. The second is to determine how pre-processing and post-processing techniques affect OCR accuracy. For The third goal is to assess and recommend improvements to the recognition accuracy of Tesseract OCR for Pesantren books. These goals and objectives guide this research as a whole, providing a structured approach to addressing the challenges and opportunities of digitizing Pesantren literature. The research entails gathering a diverse set of Pesantren books that represent various styles, fonts, and layouts commonly found for data collection. This dataset serves as the foundation for experimentation. The technical implementation entails incorporating the Tesseract OCR into the system and configuring it to recognize Arabic scripts, ensuring compatibility with the diverse content of Pesantren books. They are also applied to images of Pesantren book pages for pre-processing techniques such as binarization and noise removal. In addition, post-processing includes the implementation of a spell correction module. This module corrects OCR errors and improves the accuracy and readability of the extracted text. The core of the research is evaluating OCR accuracy. The study uses text similarity and WER to quantify the impact of pre-and post-processing techniques on OCR results. The data collected during the evaluation phase is analyzed to conclude the performance of system in digitizing Pesantren books. This analysis serves as the foundation for the research findings and recommendations. As a result, the research successfully implemented Tesseract OCR for Pesantren books. This entailed configuring Tesseract to recognize Arabic as a critical achievement for handling the mixed-script content commonly found in pesantren books. According to the findings, pre-processing techniques significantly improved OCR accuracy. The evaluation of OCR accuracy revealed significant improvements when pre-processing techniques were used. According to the findings, the most effective approach involved a combination of pre-processing and Tesseract OCR, which resulted in the highest text similarity and the lowest word error rate. Overall, the findings emphasize the importance of pre-and postprocessing techniques in improving OCR accuracy for Pesantren books, as well as practical approaches for digitizing this valuable cultural and educational content.\u003C\u002Fp>","https:\u002F\u002Frepository.polinema.ac.id\u002Fcollection\u002Fdetail?id=dspace000002403&year=2023&title=implementation-of-tesseract-ocr-in-pesantren-books-using-image-processing-and-spell-correction-at-ppss-nurul-huda"]