Melola Discovery

← Melola Discovery

Jurnal Tanpa peringkat SINTA

Ultra-Low-Cost Hybrid OCR–LLM Architecture for Production Grade E-KTP Extraction

Scientific Journal of Informatics · Vol. 12 No. 4 (2025) · hlm. 665–672

DOI
10.15294/sji.v12i4.38200
ISSN
2407-7658, 2460-0040
Terbit
2025-12-23
Penerbit
Universitas Negeri Semarang
Bahasa
eng

Abstrak

Purpose: The purpose of this study is to be able to avoid limitations of inexpensive ID card data extraction services and preserve privacy, which can simultaneously achieve reliable operation even under an environment with minimum infrastructure, in particular if no dependency on GPU-based servers are required. Method: The proposed approach is a microservice pipeline with three stages: (1) local lightweight pre-processing on devices, (2) Tesseract CPU-based OCR. js, (3) fast text tokenization through a small premature external LLM. The system is developed as TypeScript backend utilizing the Hono framework with all image processing taking place locally in order to keeping user data private. Result: The result of the experimental evaluations with real ID card samples is that the system can run stably in low-performance VPS (1 vCPU, 1 GB RAM) with operation cost approximately IDR 2.5047 per extraction process and its accuracy level is acceptable for use in a production environment. Moreover, the results indicate that system latency is dominated by LLM inference at the cloud. Novelty: The main contribution and novelty of this study is that we demonstrate, for the first time, a cost-effective (privacy-preserving) OCR-LLM hybrid pipeline without demanding expensive GPU models at large scale which makes our system suitable under limited storage and resource constraints on-premises or edge environments in small organizations including micro-SaaS services.

Kata kunci

Jaringan kolaborasi para penulis

6 penulis · 7 tautan kepenulisan bersama · 1 lintas organisasi

Arahkan kursor atau ketuk sebuah nama untuk menyorot rekan penulisnya. Klik dua kali untuk membuka halaman penulis.

  • Universitas PGRI Semarang
  • Universitas Dian Nuswantoro
  • Organisasi berbeda
  • Organisasi sama
  • Afiliasi tidak diketahui
  • Garis lebih tebal = lebih banyak artikel bersama · lingkaran lebih besar = lebih banyak artikel
Tampilkan sebagai tabel
Rekan penulisOrganisasiArtikel bersama
Danu Candra SaputraUniversitas PGRI Semarang1
Noora Qotrun NadaUniversitas PGRI Semarang1
Vilda Ana Veria SetyawatiUniversitas Dian Nuswantoro1

Afiliasi berasal dari OpenAlex; metadata OAI-PMH jurnal tidak memuatnya. Penulis tanpa ID OpenAlex dicocokkan berdasarkan nama, sehingga dua orang bisa tergabung atau satu orang bisa terpisah.

Sumber metadata: OAI-PMH (metadata jurnal) https://journal.unnes.ac.id/journals/sji/oai, pertama di-harvest 6 Nov 2026, terakhir terlihat 8 Nov 2026.
Metadata asli (oai_dc)
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en">Ultra-Low-Cost Hybrid OCR–LLM Architecture for Production Grade E-KTP Extraction</dc:title>
	<dc:creator xml:lang="en">Saputro, Anjar Tiyo</dc:creator>
	<dc:creator xml:lang="en">Herlambang, Bambang Agus</dc:creator>
	<dc:creator xml:lang="en">Novita, Mega</dc:creator>
	<dc:subject xml:lang="en">e-KTP Extraction</dc:subject>
	<dc:subject xml:lang="en">Optical character recognition</dc:subject>
	<dc:subject xml:lang="en">Large language model</dc:subject>
	<dc:subject xml:lang="en">Microservices</dc:subject>
	<dc:subject xml:lang="en">Privacy-Preserving System</dc:subject>
	<dc:description xml:lang="en">Purpose: The purpose of this study is to be able to avoid limitations of inexpensive ID card data extraction services and preserve privacy, which can simultaneously achieve reliable operation even under an environment with minimum infrastructure, in particular if no dependency on GPU-based servers are required.
Method: The proposed approach is a microservice pipeline with three stages: (1) local lightweight pre-processing on devices, (2) Tesseract CPU-based OCR. js, (3) fast text tokenization through a small premature external LLM. The system is developed as TypeScript backend utilizing the Hono framework with all image processing taking place locally in order to keeping user data private.
Result: The result of the experimental evaluations with real ID card samples is that the system can run stably in low-performance VPS (1 vCPU, 1 GB RAM) with operation cost approximately IDR 2.5047 per extraction process and its accuracy level is acceptable for use in a production environment. Moreover, the results indicate that system latency is dominated by LLM inference at the cloud.
Novelty: The main contribution and novelty of this study is that we demonstrate, for the first time, a cost-effective (privacy-preserving) OCR-LLM hybrid pipeline without demanding expensive GPU models at large scale which makes our system suitable under limited storage and resource constraints on-premises or edge environments in small organizations including micro-SaaS services.</dc:description>
	<dc:publisher xml:lang="en">Universitas Negeri Semarang</dc:publisher>
	<dc:date>2025-12-23</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:type xml:lang="en">Peer-reviewed Article</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://journal.unnes.ac.id/journals/sji/article/view/38200</dc:identifier>
	<dc:identifier>10.15294/sji.v12i4.38200</dc:identifier>
	<dc:source xml:lang="en">Scientific Journal of Informatics; Vol. 12 No. 4: Special Issue; 665-672</dc:source>
	<dc:source>2407-7658</dc:source>
	<dc:source>2460-0040</dc:source>
	<dc:source>10.15294/sji.v12i4</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://journal.unnes.ac.id/journals/sji/article/view/38200/7748</dc:relation>
	<dc:rights xml:lang="en">Copyright (c) 2025 Scientific Journal of Informatics</dc:rights>
</oai_dc:dc>