InscriptionOCR: A Dataset and Method for Understanding Inscriptions

Jaidev Sanjay Khalane; Akbar Ali; V. N. Prabhakar; Shanmuganathan Raman

cs.CVOct 7, 2026

InscriptionOCR: A Dataset and Method for Understanding Inscriptions

Authors: Jaidev Sanjay Khalane, Akbar Ali, V. N. Prabhakar, Shanmuganathan Raman

Organizations: Indian Institute of Technology Gandhinagar, India.

Abstract

Ancient script image restoration is a fundamental problem in computer vision, as it directly affects the reliable analysis and interpretation of historical documents and inscriptions. Ashokan Brahmi is an ancient script extensively used during the reign of Emperor Ashoka in the 3rd century BC, primarily for inscriptions in Prakrit. These inscriptions, including major and minor rock and pillar edicts, constitute a valuable yet largely unexplored source of data for computational analysis. The degraded nature of inscription imagery and the lack of standardized digital resources pose significant challenges for automated processing. We present an end-to-end AI-based framework for understanding ancient inscriptions that encompasses image enhancement, optical character recognition (OCR), transliteration, and neural machine translation (NMT). The proposed pipeline processes low-quality images captured directly from stone inscriptions, performs image restoration and Brahmi script character recognition, maps the recognized characters to the Roman script, and finally translates the resulting Prakrit text into English. We also introduce two new datasets: (i) InscriptionOCR Dataset: the largest publicly usable digital OCR dataset for Brahmi script to date, consisting of over 200,000 character images across about 600 classes, and (ii) a bilingual Prakrit-English parallel corpus comprising over 2,000 sentence pairs for NMT. We believe that the proposed framework and datasets will facilitate future research in ancient script analysis, low-resource OCR, and digital epigraphy.

Figures & tables

Figure 1 : Representative dataset characters.

Author	Translation	Dataset Size	OCR Classes	Translation Dataset
Gautam [ 7 ]	×	7,011	170	×
Gunasekara [ 9 ]	×	250–300	×	×
Mubarakkaa [ 13 ]	×	×	×	×
Dhivya [ 4 ]	×	190,000	209	×
Chaudhari [ 3 ]	✓	×	×	1474
Our Work	✓	253,800	564	2224

Table 1 : Comparison of OCR and translation capabilities for Brahmi-based datasets.

OCR Dataset
Total Images	253800
Training Images	228,420
Testing Images	25,380
OCR Class-wise Distribution
Numeric Characters	30
Vowel Characters	17

Table 2 : Specification of our proposed dataset.

Figure 2 : The Stepwise Results in Image Enhancement Phase.

Figure 3 : The figure illustrates the performance of the proposed method. Best viewed when zoomed in .

Parameter	Value
Batch Size	32
Optimizer	Adam (Adaptive Moment Estimation)
Learning Rate	0.001
Loss Function	Categorical Cross-Entropy
Number of Epochs	20

Table 3 : CNN training parameters.

Parameter for NMT Model Finetuning	Value
Learning Rate	$5\times 10^{-5}$
Batch Size	8
Weight Decay	0.01
Number of Training Epochs	3

Table 4 : NMT fine-tuning parameters.

Figure 4 : Overall flow in our Inscription Framework.

Model	Accuracy	Precision	Recall	F1 Score
LeNet-5	0.8872	0.8912	0.8872	0.8753
VGG-16	0.9939	0.9955	0.9939	0.9935
EfficientNet	0.9996	0.9997	0.9996	0.9996
ResNet-50	1.0000	1.0000	1.0000	1.0000

Table 5 : Performance comparison of CNN-based OCR models.

Model Name	BLEU Score	METEOR Score
M2M 100 418M	56.367	0.772
t5 Base	7.750	0.093
Helsinki-NLP Opus	14.120	0.254

Table 6 : NMT training results.

Panel of Figure 5 — Figure 5 : Comparison of the Original Image and Image Enhancement Results

Panel of Figure 6 — Figure 6 : OCR Processing Pipeline for Inscription Images

Panel of Figure 7 — Figure 7 : Image Enhancement Application Interface and Output Result

Panel of Figure 8 — Figure 8 : Image Annotation Process and Results

Panel of Figure 9 — Figure 9 : Mobile Application Interface and OCR+Transliteration+NMT Output

Explore similar work

Apr 19, 2026cs.CV

MESA: A Training-Free Multi-Exemplar Deep Framework for Restoring Ancient Inscription Textures

Ancient inscriptions frequently suffer missing or corrupted regions from fragmentation, erosion, or other damage, hindering reading, and analysis. We review prior image restoration methods and their applicability to inscription image recovery, then introduce MESA (Multi-Exemplar, Style-Aware) -an image-level restoration method that uses well-preserved exemplar inscriptions (from the same epigraphic monument, material, or similar letterforms) to guide reconstruction of damaged text. MESA encodes VGG19 convolutional features as Gram matrices to capture exemplar texture, style, and stroke structure; for each neural network layer it selects the exemplar minimizing Mean-Squared Displacement (MSD) to the damaged input. Layer-wise contribution weights are derived from Optical Character Recognition-estimated character widths in the exemplar set to bias filters toward scales matching letter geometry, and a training mask preserves intact regions so synthesis is restricted to damaged areas. We also summarize prior network architectures and exemplar and single-image synthesis, inpainting, and Generative Adversarial Network (GAN) approaches, highlighting limitations that MESA addresses. Comparative experiments demonstrate the advantages of MESA. Finally, we provide a practical roadmap for choosing restoration strategies given available exemplars and metadata.

Vasileios Toulatzis, Sofia Theodoridou, Ioannis Fudos

University of Ioannina, Greece

Image Generation Image Restoration Image Inpainting Texture Synthesis

Ask PDF

Sep 15, 2026cs.CV

FROD: Feature Matching Residual Denoising Oracle Bone Decipher

Oracle bone script (OBS), one of the earliest Chinese writing systems, plays an important role in the study of Chinese etymology. Traditional decipherment relies heavily on domain experts who analyze characters through semantic context and structural evolution. To assist this labor-intensive process, we formulate OBS decipherment assistance as a cross-era image translation task and propose FROD (Feature Matching Residual Denoising Oracle Bone Decipher). Although many OBS characters differ substantially from their modern counterparts, they often preserve local topological invariants at the radical level. During training, FROD leverages fast feature matching to provide gated segmentation supervision: paired samples with sufficient matches are processed patch-wise to align fine-grained radicals, whereas low-similarity pairs are trained holistically to avoid mismatched artifacts. In addition, a Residual Denoising Diffusion Model (RDDM) jointly estimates noise and residual signals, thereby reducing the positional drift and stroke disorder commonly observed in standard diffusion models. Finally, a multi-stage font stylization refinement network refines the generated images by eliminating edge noise and stabilizing stroke structures. On our augmented character-disjoint dataset, FROD achieves higher Top-1 recognition accuracy than the evaluated baselines, with a 3.8% absolute gain over OBSD.

Yanbin Hou, Biao Xiong, Guojun Xu +4

School of Artificial Intelligence, Wuhan University of Technology, Wuhan 430070, China

Image-to-Image Translation Diffusion Models

Ask PDF

Jul 20, 2026cs.HC

AlphaOracle: Oracle bone script decipherment via human-workflow-inspired deep learning

Approximately 3,000 of the 4,500 oracle bone script (OBS) characters remain undeciphered due to fragmentary inscriptions and sparse evidence. Current AI approaches fail to replicate expert workflows that integrate form analysis, contextual semantics, and philological reasoning. We introduce AlphaOracle, a human-workflow-inspired framework that systematizes OBS decipherment using the largest digitized corpus to date. Its multi-stage pipeline comprises: (i) rubbing parsing; (ii) radical-based morphological analysis with diachronic modeling; (iii) contextual retrieval with semantic alignment; and (iv) philological validation against classical sources. Each stage generates explicit, confidence-weighted evidence chains, culminating in interpretable reports for scholarly verification. Across multiple test characters, AlphaOracle's readings strongly agreed with expert interpretations. In a study of 86 domain specialists, it reduced analysis time by 64% and 79% of participants rated it highly useful. Notably, AlphaOracle resolves the character "Lao" as a toponymic or clan designation, offering concrete revisions to Shang administrative and social interpretations. These results suggest that computational methods aligned with philological practice can facilitate OBS research and provide a conceptual reference for studies of other undeciphered scripts.

Yuliang Liu, Haisu Guan, Pengjie Wang +11

School of Software Engineering, Huazhong University of Science and Technology, Wuhan 430074, China · Key Laboratory of Oracle Bone Inscriptions Information Processing, Anyang Normal University, Anyang 455002, China · School of Electronic Information Engineering, South China University of Technology, Guangzhou 510641, China

Topic pending

Ask PDF