Optical_Character
-
KrishokBondhu: A Retrieval-Augmented Voice-Based Agricultural Advisory Call Center for Bengali Farmers
2025-10-21 07:24:55Mohd Ruhul Ameen, Akif Islam, Farjana Aktar, M. Saifuzzaman RafatAbstract
In Bangladesh, many farmers continue to face challenges in accessing timely, expert-level agricultural guidance. This paper presents KrishokBondhu, a voice-enabled, call-centre-integrated advisory platform built on a Retrieval-Augmented Generation (RAG) framework, designed specifically for Bengali-speaking farmers. The system aggregates authoritative agricultural handbooks, extension manuals, and NGO publications; applies Optical Character Recognition (OCR) and document-parsing pipelines to digitize and structure the content; and indexes this corpus in a vector database for efficient semantic retrieval. Through a simple phone-based interface, farmers can call the system to receive real-time, context-aware advice: speech-to-text converts the Bengali query, the RAG module retrieves relevant content, a large language model (Gemma 3-4B) generates a context-grounded response, and text-to-speech delivers the answer in natural spoken Bengali. In a pilot evaluation, KrishokBondhu produced high-quality responses for 72.7% of diverse agricultural queries covering crop management, disease control, and cultivation practices. Compared to the KisanQRS benchmark, the system achieved a composite score of 4.53 (vs. 3.13) on a 5-point scale, a 44.7% improvement, with especially large gains in contextual richness (+367%) and completeness (+100.4%), while maintaining comparable relevance and technical specificity. Semantic similarity analysis further revealed a strong correlation between retrieved context and answer quality, emphasizing the importance of grounding generative responses in curated documentation. KrishokBondhu demonstrates the feasibility of integrating call-centre accessibility, multilingual voice interaction, and modern RAG techniques to deliver expert-level agricultural guidance to remote Bangladeshi farmers, paving the way toward a fully AI-driven agricultural advisory ecosystem.
Abstract (translated)
在孟加拉国，许多农民仍然面临着难以获得及时、专家级别的农业指导的挑战。本文介绍了一种名为KrishokBondhu的声音驱动、呼叫中心集成咨询平台，该平台专为说孟加拉语的农民设计，并基于检索增强生成（RAG）框架构建。系统汇集了权威性的农业手册、推广指南和非政府组织出版物；使用光学字符识别（OCR）和文档解析管道将内容数字化并结构化；并将这些信息存储在一个向量数据库中，以便高效地进行语义检索。通过一个简单的电话界面，农民可以通过拨打该系统来获取实时的上下文感知建议：语音转文本技术会将孟加拉语查询转换为文本，RAG模块检索相关的内容，一个大型语言模型（Gemma 3-4B）生成基于上下文的回答，并以自然口语的形式通过文字转语音技术用孟加拉语传达答案。在试点评估中，KrishokBondhu对涵盖了作物管理、病害控制和栽培实践等多方面农业问题的广泛查询，产生了高质量回答的比例达到了72.7%。与KisanQRS基准系统相比，在5分制评分体系下该系统的综合评分为4.53（而KisanQRS为3.13），得分提高了44.7%，特别是在上下文丰富度（+367%）和完整性（+100.4%）方面有显著提升，同时保持了相当的关联性和技术精确性。语义相似性分析进一步揭示了检索到的内容与答案质量之间的强烈相关性，强调了基于精心整理文档生成回应的重要性。KrishokBondhu展示了将呼叫中心可访问性、多语言语音交互和现代RAG技术整合起来为偏远地区的孟加拉国农民提供专家级农业指导的可行性，并为全由AI驱动的农业咨询生态系统铺平了道路。
URL
https://arxiv.org/abs/2510.18355
PDF
-
CharDiff: A Diffusion Model with Character-Level Guidance for License Plate Image Restoration
2025-10-20 09:23:29Gyuhwan Park, Kihyun Na, Injung KimAbstract
The significance of license plate image restoration goes beyond the preprocessing stage of License Plate Recognition (LPR) systems, as it also serves various purposes, including increasing evidential value, enhancing the clarity of visual interface, and facilitating further utilization of license plate images. We propose a novel diffusion-based framework with character-level guidance, CharDiff, which effectively restores and recognizes severely degraded license plate images captured under realistic conditions. CharDiff leverages fine-grained character-level priors extracted through external segmentation and Optical Character Recognition (OCR) modules tailored for low-quality license plate images. For precise and focused guidance, CharDiff incorporates a novel Character-guided Attention through Region-wise Masking (CHARM) module, which ensures that each character's guidance is restricted to its own region, thereby avoiding interference with other regions. In experiments, CharDiff significantly outperformed the baseline restoration models in both restoration quality and recognition accuracy, achieving a 28% relative reduction in CER on the Roboflow-LP dataset, compared to the best-performing baseline model. These results indicate that the structured character-guided conditioning effectively enhances the robustness of diffusion-based license plate restoration and recognition in practical deployment scenarios.
Abstract (translated)
车牌图像恢复的意义不仅限于车牌识别（LPR）系统的预处理阶段，还服务于多种目的，包括增加证据价值、提高视觉界面的清晰度和促进进一步利用车牌图像。我们提出了一种基于扩散的新框架，并结合字符级指导——CharDiff。该方法有效地修复并识别在现实条件下捕捉到的严重退化的车牌图像。 CharDiff 利用通过外部分割模块及专为低质量车牌图像设计的光学字符识别（OCR）模块提取出的精细粒度字符先验知识，从而实现高效恢复和识别功能。为了提供精确且集中的指导，CharDiff 集成了一种创新性的基于区域掩码的字符引导注意力机制（CHARM）模块，确保每个字符的指导仅限于其自身的区域内，避免对其他区域产生干扰。 实验结果显示，在图像修复质量和识别准确性方面，CharDiff 显著优于基准恢复模型。在Roboflow-LP数据集上，与表现最佳的基准模型相比，CharDiff 将CER相对降低了28%。这些结果表明，结构化的字符引导调节有效提升了基于扩散的车牌恢复和识别方法在实际部署场景中的鲁棒性。
URL
https://arxiv.org/abs/2510.17330
PDF
-
APGNet: Adaptive Prior-Guided for Underwater Camouflaged Object Detection
2025-10-14 01:51:44Xinxin Huang, Han Sun, Junmin Cai, Ningzhong Liu, Huiyu ZhouAbstract
Detecting camouflaged objects in underwater environments is crucial for marine ecological research and resource exploration. However, existing methods face two key challenges: underwater image degradation, including low contrast and color distortion, and the natural camouflage of marine organisms. Traditional image enhancement techniques struggle to restore critical features in degraded images, while camouflaged object detection (COD) methods developed for terrestrial scenes often fail to adapt to underwater environments due to the lack of consideration for underwater optical characteristics. To address these issues, we propose APGNet, an Adaptive Prior-Guided Network, which integrates a Siamese architecture with a novel prior-guided mechanism to enhance robustness and detection accuracy. First, we employ the Multi-Scale Retinex with Color Restoration (MSRCR) algorithm for data augmentation, generating illumination-invariant images to mitigate degradation effects. Second, we design an Extended Receptive Field (ERF) module combined with a Multi-Scale Progressive Decoder (MPD) to capture multi-scale contextual information and refine feature representations. Furthermore, we propose an adaptive prior-guided mechanism that hierarchically fuses position and boundary priors by embedding spatial attention in high-level features for coarse localization and using deformable convolution to refine contours in low-level features. Extensive experimental results on two public MAS datasets demonstrate that our proposed method APGNet outperforms 15 state-of-art methods under widely used evaluation metrics.
Abstract (translated)
在水下环境中检测伪装物体对于海洋生态研究和资源勘探至关重要。然而，现有方法面临着两个主要挑战：水中图像退化（包括低对比度和颜色失真）以及海洋生物的天然伪装。传统图像增强技术难以恢复受损图像中的关键特征，而为陆地场景开发的伪装物体检测(COD) 方法由于缺乏对水下光学特性的考虑，往往无法适应水下环境。为了应对这些问题，我们提出了一种自适应先验引导网络（APGNet），它集成了孪生架构与新颖的先验指导机制来提高鲁棒性和检测精度。 首先，我们采用多尺度Retinex彩色恢复（MSRCR）算法进行数据增强，生成照明不变图像以减轻退化效应。其次，我们设计了一个扩展感受野（ERF）模块结合了多尺度渐进解码器（MPD），用于捕捉多尺度上下文信息并细化特征表示。此外，我们还提出了一种自适应先验引导机制，通过在高级别特征中嵌入空间注意来进行位置和边界先验的层次融合，进行粗定位，并使用可变形卷积来精细低级别特征中的轮廓。 在两个公开的MAS数据集上的大量实验结果表明，在广泛使用的评估指标下，我们提出的方法APGNet优于15种最先进的方法。
URL
https://arxiv.org/abs/2510.12056
PDF
-
A Large-Language-Model Assisted Automated Scale Bar Detection and Extraction Framework for Scanning Electron Microscopic Images
2025-10-13 10:50:54Yuxuan Chen, Ruotong Yang, Zhengyang Zhang, Mehreen Ahmed, Yanming WangAbstract
Microscopic characterizations, such as Scanning Electron Microscopy (SEM), are widely used in scientific research for visualizing and analyzing microstructures. Determining the scale bars is an important first step of accurate SEM analysis; however, currently, it mainly relies on manual operations, which is both time-consuming and prone to errors. To address this issue, we propose a multi-modal and automated scale bar detection and extraction framework that provides concurrent object detection, text detection and text recognition with a Large Language Model (LLM) agent. The proposed framework operates in four phases; i) Automatic Dataset Generation (Auto-DG) model to synthesize a diverse dataset of SEM images ensuring robust training and high generalizability of the model, ii) scale bar object detection, iii) information extraction using a hybrid Optical Character Recognition (OCR) system with DenseNet and Convolutional Recurrent Neural Network (CRNN) based algorithms, iv) an LLM agent to analyze and verify accuracy of the results. The proposed model demonstrates a strong performance in object detection and accurate localization with a precision of 100%, recall of 95.8%, and a mean Average Precision (mAP) of 99.2% at IoU=0.5 and 69.1% at IoU=0.5:0.95. The hybrid OCR system achieved 89% precision, 65% recall, and a 75% F1 score on the Auto-DG dataset, significantly outperforming several mainstream standalone engines, highlighting its reliability for scientific image analysis. The LLM is introduced as a reasoning engine as well as an intelligent assistant that suggests follow-up steps and verifies the results. This automated method powered by an LLM agent significantly enhances the efficiency and accuracy of scale bar detection and extraction in SEM images, providing a valuable tool for microscopic analysis and advancing the field of scientific imaging.
Abstract (translated)
微观表征技术，如扫描电子显微镜（SEM），在科学研究中广泛用于可视化和分析微观结构。确定标尺是准确进行SEM分析的重要第一步；然而，目前主要依靠手动操作来完成这一过程，这不仅耗时而且容易出错。为了解决这个问题，我们提出了一种多模态且自动化的标尺检测与提取框架，该框架结合了对象检测、文本检测和基于大型语言模型（LLM）的文本识别功能。 提出的框架分为四个阶段：i) 自动数据生成(Auto-DG)模型，用于合成各种SEM图像的数据集，确保模型训练的稳健性和高泛化能力；ii) 标尺对象检测；iii) 使用密集网络(DenseNet)和卷积循环神经网络(CRNN)算法组成的混合光学字符识别(OCR)系统进行信息提取；iv) 一个LLM代理来分析并验证结果的准确性。 所提出的模型在物体检测方面表现出色，且精确定位精度为100%，召回率为95.8%，以IoU=0.5时mAP值为99.2%，IoU=0.5:0.95范围内的mAP值为69.1%。混合OCR系统在Auto-DG数据集上实现了89%的精度、65%的召回率和75%的F1分数，显著优于多个主流独立引擎的表现，证明其对于科学图像分析的可靠性。 大型语言模型不仅作为推理引擎工作，还充当智能助手提出后续步骤并验证结果。这种方法通过LLM代理的自动化手段，在SEM图像中显著提高了标尺检测与提取的效率和准确性，为微观结构分析提供了一个有价值的工具，并推动了科学成像领域的发展。
URL
https://arxiv.org/abs/2510.11260
PDF
-
VOLTAGE: A Versatile Contrastive Learning based OCR Methodology for ultra low-resource scripts through Auto Glyph Feature Extraction
2025-10-12 07:47:41Prawaal Sharma, Poonam Goyal, Vidisha Sharma, Navneet GoyalAbstract
UNESCO has classified 2500 out of 7000 languages spoken worldwide as endangered. Attrition of a language leads to loss of traditional wisdom, folk literature, and the essence of the community that uses it. It is therefore imperative to bring digital inclusion to these languages and avoid its extinction. Low resource languages are at a greater risk of extinction. Lack of unsupervised Optical Character Recognition(OCR) methodologies for low resource languages is one of the reasons impeding their digital inclusion. We propose VOLTAGE - a contrastive learning based OCR methodology, leveraging auto-glyph feature recommendation for cluster-based labelling. We augment the labelled data for diversity and volume using image transformations and Generative Adversarial Networks. Voltage has been designed using Takri - a family of scripts used in 16th to 20th century in the Himalayan regions of India. We present results for Takri along with other Indic scripts (both low and high resource) to substantiate the universal behavior of the methodology. An accuracy of 95% for machine printed and 87% for handwritten samples on Takri script has been achieved. We conduct baseline and ablation studies along with building downstream use cases for Takri, demonstrating the usefulness of our work.
Abstract (translated)
联合国教科文组织将全球约7000种语言中的2500种列为濒危语言。一种语言的衰落会导致传统智慧、民间文学以及使用该语言的社区本质的丧失。因此，有必要为这些语言带来数字包容性，以避免其灭绝。资源较少的语言面临更高的灭绝风险。缺乏针对低资源语言的无监督光学字符识别（OCR）方法是阻碍它们实现数字包容性的原因之一。我们提出了一种基于对比学习的方法——VOLTAGE，该方法利用自动字形特征推荐进行基于聚类的标签生成，并通过图像变换和生成对抗网络来扩充标注数据集的多样性和规模。VOLTAGE使用印度喜马拉雅地区16至20世纪使用的Takri文字家族设计开发而成。我们提供了Takri以及其他印地语（包括资源较少和较多的语言）的实验结果，以证明该方法具有普遍适用性。在Takri字符上，机器打印样本和手写样本分别达到了95%和87%的准确率。我们还进行了基线测试、消融研究，并构建了Takri语言下游应用场景，以展示我们的工作的实用性。
URL
https://arxiv.org/abs/2510.10490
PDF
-
Detecting Legend Items on Historical Maps Using GPT-4o with In-Context Learning
2025-10-09 16:08:48Sofia Kirsanova, Yao-Yi Chiang, Weiwei DuanAbstract
Historical map legends are critical for interpreting cartographic symbols. However, their inconsistent layouts and unstructured formats make automatic extraction challenging. Prior work focuses primarily on segmentation or general optical character recognition (OCR), with few methods effectively matching legend symbols to their corresponding descriptions in a structured manner. We present a method that combines LayoutLMv3 for layout detection with GPT-4o using in-context learning to detect and link legend items and their descriptions via bounding box predictions. Our experiments show that GPT-4 with structured JSON prompts outperforms the baseline, achieving 88% F-1 and 85% IoU, and reveal how prompt design, example counts, and layout alignment affect performance. This approach supports scalable, layout-aware legend parsing and improves the indexing and searchability of historical maps across various visual styles.
Abstract (translated)
历史地图图例对于解读地图符号至关重要。然而，由于其布局不一致和格式无结构化，自动提取这些信息具有挑战性。先前的工作主要集中在分割或通用光学字符识别（OCR）上，而很少有方法能够以结构化方式有效匹配图例符号及其相应的描述。我们提出了一种结合LayoutLMv3进行版面检测与使用GPT-4和上下文学习通过边界框预测来检测并链接图例项及其描述的方法。我们的实验表明，在使用结构化的JSON提示时，GPT-4的表现优于基线模型，F1分数达到88%，IoU（交并比）为85%。此外，这些实验揭示了提示设计、示例数量以及布局对齐如何影响性能。这种方法支持大规模且版面感知的图例解析，并改善了各种视觉风格下历史地图的索引和可搜索性。
URL
https://arxiv.org/abs/2510.08385
PDF
-
Look before Transcription: End-to-End SlideASR with Visually-Anchored Policy Optimization
2025-10-08 08:18:47Rui Hu, Delai Qiu, Yining Wang, Shengping Liu, Jitao SangAbstract
Automatic speech recognition (ASR) systems often struggle with domain-specific terminology, especially in specialized settings such as academic lectures. To address this, we define the SlideASR task, which leverages the rich visual information from presentation slides to improve transcription accuracy. Existing pipeline methods for this task tend to be complex and underperform. Although omni-modal large language models (OLLMs) provide a promising end-to-end framework, they frequently fail in practice by degenerating into simple optical character recognition (OCR) systems. To overcome this, we propose Visually-Anchored Policy Optimization (VAPO), a novel post-training method designed to control the model's reasoning process. Drawing on the Chain-of-Thought reasoning paradigm, VAPO enforces a structured "Look before Transcription" procedure using a <think><answer> format. Specifically, the model first performs OCR on the slide content within the think step, then generates the transcription by referencing this recognized visual information in the answer step. This reasoning process is optimized via reinforcement learning with four distinct rewards targeting format compliance, OCR accuracy, ASR quality, and visual anchoring consistency. To support further research, we construct SlideASR-Bench, a new entity-rich benchmark consisting of a synthetic dataset for training and testing, and a challenging real-world set for evaluation. Extensive experiments demonstrate that VAPO significantly improves recognition of domain-specific terms, establishing an effective end-to-end paradigm for SlideASR.
Abstract (translated)
自动语音识别（ASR）系统在处理特定领域的术语时常常遇到困难，尤其是在学术讲座等专业环境中。为了解决这个问题，我们定义了SlideASR任务，该任务利用演示幻灯片中的丰富视觉信息来提高转录准确性。现有的管道方法对于此任务通常很复杂且表现不佳。虽然全能型大型语言模型（OLLM）提供了一个有前途的端到端框架，但它们在实践中往往会退化为简单的光学字符识别（OCR）系统而失败。为了克服这个问题，我们提出了视觉锚定策略优化（VAPO），这是一种新颖的训练后方法，旨在控制模型的推理过程。借鉴链式思维推理范式，VAPO通过采用<think><answer>格式强制执行结构化的“先查看再转录”程序。具体来说，在思考步骤中，模型首先对幻灯片内容进行OCR处理，然后在回答步骤中引用已识别的视觉信息生成转录文本。该推理过程通过强化学习优化，并针对格式合规性、OCR准确性、ASR质量和视觉锚定一致性设置了四种不同的奖励。 为了支持进一步的研究，我们构建了SlideASR-Bench，这是一个新的实体丰富的基准测试，包括用于训练和测试的人工合成数据集以及一个具有挑战性的现实世界评估数据集。广泛的实验表明，VAPO显著提高了特定领域术语的识别能力，并为SlideASR建立了有效的端到端范式。
URL
https://arxiv.org/abs/2510.08618
PDF
-
Evaluating OCR performance on food packaging labels in South Africa
2025-10-03 23:38:45Mayimunah Nagayi, Alice Khan, Tamryn Frank, Rina Swart, Clement NyirendaAbstract
This study evaluates four open-source Optical Character Recognition (OCR) systems which are Tesseract, EasyOCR, PaddleOCR, and TrOCR on real world food packaging images. The aim is to assess their ability to extract ingredient lists and nutrition facts panels. Accurate OCR for packaging is important for compliance and nutrition monitoring but is challenging due to multilingual text, dense layouts, varied fonts, glare, and curved surfaces. A dataset of 231 products (1,628 images) was processed by all four models to assess speed and coverage, and a ground truth subset of 113 images (60 products) was created for accuracy evaluation. Metrics include Character Error Rate (CER), Word Error Rate (WER), BLEU, ROUGE-L, F1, coverage, and execution time. On the ground truth subset, Tesseract achieved the lowest CER (0.912) and the highest BLEU (0.245). EasyOCR provided a good balance between accuracy and multilingual support. PaddleOCR achieved near complete coverage but was slower because it ran on CPU only due to GPU incompatibility, and TrOCR produced the weakest results despite GPU acceleration. These results provide a packaging-specific benchmark, establish a baseline, and highlight directions for layout-aware methods and text localization.
Abstract (translated)
这项研究评估了四种开源光学字符识别（OCR）系统——Tesseract、EasyOCR、PaddleOCR和TrOCR在真实世界食品包装图像上的表现。该研究的目的是评估这些系统提取成分列表和营养信息面板的能力。对于包装上的准确OCR来说，合规性和营养监测非常重要，但由于多语言文本、密集布局、不同字体、反光以及曲面等因素的存在，这一任务具有挑战性。 为了评估速度和覆盖率，所有四个模型都处理了一个包含231种产品（共1,628张图像）的数据集，并为准确性评估创建了113张图像（来自60种产品）的基准数据。评估指标包括字符错误率（CER）、单词错误率（WER）、BLEU、ROUGE-L、F1分数、覆盖率和执行时间。 在基准数据子集中，Tesseract实现了最低的字符错误率（CER为0.912）和最高的BLEU得分（0.245）。EasyOCR在准确性和多语言支持之间提供了良好的平衡。PaddleOCR达到了几乎完全的覆盖范围，但由于与GPU不兼容而仅运行于CPU上，因此速度较慢。尽管有GPU加速，但TrOCR的表现最弱。 这些结果为包装特定领域的基准测试奠定了基础，并突显了布局感知方法和文本定位方向的重要性。
URL
https://arxiv.org/abs/2510.03570
PDF
-
Interpret, prune and distill Donut : towards lightweight VLMs for VQA on document
2025-09-30 13:31:03Adnan Ben Mansour, Ayoub Karine, David NaccacheAbstract
Recent advances in Visually-rich Document Understanding rely on large Vision-Language Models like Donut, which perform document-level Visual Question Answering without Optical Character Recognition. Despite their effectiveness, these models are too costly for real-time or resource-constrained applications. We investigate model compression through knowledge distillation, training compact student models from a larger teacher. We leverage mechanistic interpretability to drive student architecture design within this framework. By analyzing internal computations, we identify essential subcomponents to retain, while having a clear view of which subcomponents should be approximated, skipped, or reparametrized based on their function. This approach yields Donut-MINT (Mechanistic Interpretability-based Network Trimming), a pruned Donut variant that reduces inference time and memory usage while maintaining strong performance on DocVQA, a standard benchmark for document Visual Question Answering. Our method reframes compression as circuit discovery, bridging interpretability research and practical Vision-Language Model deployment.
Abstract (translated)
近期在丰富视觉文档理解方面取得的进展依赖于像Donut这样的大型视觉-语言模型，这些模型能够在没有光学字符识别的情况下执行基于文档级别的视觉问答任务。尽管这些模型非常有效，但它们对于实时应用或资源受限的应用来说成本过高。我们通过知识蒸馏进行模型压缩的研究，即从一个较大的教师模型训练出更加紧凑的学生模型，并利用机制性可解释性来指导学生网络架构的设计。通过分析内部计算过程，我们可以识别出需要保留的关键子组件，同时明确哪些子组件应该被近似、省略或重新参数化，这取决于它们的功能。 这种策略产生了Donut-MINT（基于机制性可解释性的网络修剪），这是一种经过优化的Donut变体，它在减少推理时间和内存使用的同时，在文档视觉问答的标准基准DocVQA上保持了强大的性能。我们的方法将压缩问题重新定义为电路发现，并且连接了可解释性研究和实用的视觉-语言模型部署之间的桥梁。
URL
https://arxiv.org/abs/2509.26235
PDF
-
Multi-modal Data Spectrum: Multi-modal Datasets are Multi-dimensional
2025-09-27 21:13:29Divyam Madaan, Varshan Muhunthan, Kyunghyun Cho, Sumit ChopraAbstract
Understanding the interplay between intra-modality dependencies (the contribution of an individual modality to a target task) and inter-modality dependencies (the relationships between modalities and the target task) is fundamental to advancing multi-modal learning. However, the nature of and interaction between these dependencies within current benchmark evaluations remains poorly characterized. In this work, we present a large-scale empirical study to quantify these dependencies across 23 visual question-answering benchmarks using multi-modal large language models (MLLMs) covering domains such as general and expert knowledge reasoning, optical character recognition, and document understanding. Our findings show that the reliance on vision, question (text), and their interaction varies significantly, both across and within benchmarks. We discover that numerous benchmarks intended to mitigate text-only biases have inadvertently amplified image-only dependencies. This characterization persists across model sizes, as larger models often use these intra-modality dependencies to achieve high performance that mask an underlying lack of multi-modal reasoning. We provide a quantitative characterization of multi-modal datasets, enabling a principled approach to multi-modal benchmark design and evaluation.
Abstract (translated)
理解模态内部依赖（即单一模态对目标任务的贡献）与跨模态依赖（即不同模态之间以及它们与目标任务之间的关系）对于推进多模态学习至关重要。然而，当前基准评估中这些依赖性及其相互作用的本质仍然未被充分描述。在这项工作中，我们进行了一次大规模实证研究，旨在通过覆盖通用和专家知识推理、光学字符识别及文档理解等领域的多模态大型语言模型（MLLMs），量化23个视觉问答基准中的这些依赖关系。我们的发现表明，在不同基准之间以及同一基准内部，对视觉信息、问题文本及其相互作用的依赖性存在显著差异。我们还发现，许多旨在缓解仅基于文本偏差的设计，反而无意中放大了图像单一模态的依赖性。这种特征在各种模型大小中都普遍存在，大型模型往往利用这些单模态依赖来实现高表现，从而掩盖其内部缺乏多模态推理能力的问题。 我们的研究提供了对多模态数据集的量化描述，为进一步制定和评估多模态基准测试提供了一个原则性的方法论。
URL
https://arxiv.org/abs/2509.23499
PDF
-
TrueGradeAI: Retrieval-Augmented and Bias-Resistant AI for Transparent and Explainable Digital Assessments
2025-09-26 16:00:36Rakesh Thakur, Shivaansh Kaushik, Gauri Chopra, Harsh RohillaAbstract
This paper introduces TrueGradeAI, an AI-driven digital examination framework designed to overcome the shortcomings of traditional paper-based assessments, including excessive paper usage, logistical complexity, grading delays, and evaluator bias. The system preserves natural handwriting by capturing stylus input on secure tablets and applying transformer-based optical character recognition for transcription. Evaluation is conducted through a retrieval-augmented pipeline that integrates faculty solutions, cache layers, and external references, enabling a large language model to assign scores with explicit, evidence-linked reasoning. Unlike prior tablet-based exam systems that primarily digitize responses, TrueGradeAI advances the field by incorporating explainable automation, bias mitigation, and auditable grading trails. By uniting handwriting preservation with scalable and transparent evaluation, the framework reduces environmental costs, accelerates feedback cycles, and progressively builds a reusable knowledge base, while actively working to mitigate grading bias and ensure fairness in assessment.
Abstract (translated)
本文介绍了TrueGradeAI，这是一个基于人工智能的数字考试框架，旨在克服传统纸质评估方法的不足之处，包括过度使用纸张、物流复杂性、评分延迟和评卷人偏见等问题。该系统通过在安全平板电脑上捕捉触控笔输入来保留自然书写，并采用基于转换器的光学字符识别技术进行转录。评估过程则通过一个增强检索功能的工作流程完成，这一工作流程整合了教师提供的答案方案、缓存层以及外部参考文献，使大型语言模型能够依据具体证据给出评分并提供明确的理由说明。 与以往仅仅数字化考生回答的平板电脑考试系统不同，TrueGradeAI在解释性自动化、偏见缓解和可追溯评分记录方面取得了进步。通过结合书写保存和规模化的透明评估方式，该框架降低了环境成本，加速了反馈周期，并逐步建立了一个可重复使用的知识库，同时积极努力减少评分偏见并确保评价的公平性。
URL
https://arxiv.org/abs/2509.22516
PDF
-
CHURRO: Making History Readable with an Open-Weight Large Vision-Language Model for High-Accuracy, Low-Cost Historical Text Recognition
2025-09-24 05:38:45Sina J. Semnani, Han Zhang, Xinyan He, Merve Tekg\"urler, Monica S. LamAbstract
Accurate text recognition for historical documents can greatly advance the study and preservation of cultural heritage. Existing vision-language models (VLMs), however, are designed for modern, standardized texts and are not equipped to read the diverse languages and scripts, irregular layouts, and frequent degradation found in historical materials. This paper presents CHURRO, a 3B-parameter open-weight VLM specialized for historical text recognition. The model is trained on CHURRO-DS, the largest historical text recognition dataset to date. CHURRO-DS unifies 155 historical corpora comprising 99,491 pages, spanning 22 centuries of textual heritage across 46 language clusters, including historical variants and dead languages. We evaluate several open-weight and closed VLMs and optical character recognition (OCR) systems on CHURRO-DS and find that CHURRO outperforms all other VLMs. On the CHURRO-DS test set, CHURRO achieves 82.3% (printed) and 70.1% (handwritten) normalized Levenshtein similarity, surpassing the second-best model, Gemini 2.5 Pro, by 1.4% and 6.5%, respectively, while being 15.5 times more cost-effective. By releasing the model and dataset, we aim to enable community-driven research to improve the readability of historical texts and accelerate scholarship.
Abstract (translated)
准确的文字识别对于历史文献的研究和文化遗产的保护具有重要意义。然而，现有的视觉-语言模型（VLMs）主要针对现代标准文本设计，无法处理历史材料中多样化的语言和文字、不规则布局以及频繁出现的质量退化问题。本文介绍了CHURRO，这是一种专为历史文本识别而设计的30亿参数开放权重视觉-语言模型（VLM）。该模型是在迄今为止最大的历史文本识别数据集CHURRO-DS上进行训练的。CHURRO-DS统一了155个历史语料库，包含99,491页文献，涵盖了跨越22世纪、涉及46种语言集群的历史文字遗产，包括历史上变异的语言和已灭绝的语言。 我们评估了几种开放权重和封闭VLM模型以及光学字符识别（OCR）系统在CHURRO-DS上的表现，并发现CHURRO的表现优于所有其他VLM。在CHURRO-DS测试集上，CHURRO在印刷文本中的标准化莱文斯坦相似度得分为82.3%，手写文本得分70.1%，分别超过第二名模型Gemini 2.5 Pro的1.4%和6.5%，并且成本效益高出15.5倍。 通过发布该模型和数据集，我们旨在推动社区驱动的研究以提高历史文献的文字可读性，并加速学术研究的发展。
URL
https://arxiv.org/abs/2509.19768
PDF
-
Logics-Parsing Technical Report
2025-09-24 04:54:37Xiangyang Chen, Shuzhao Li, Xiuwen Zhu, Yongfan Chen, Fan Yang, Cheng Fang, Lin Qu, Xiaoxiao Xu, Hu Wei, Minggang WuAbstract
Recent advances in Large Vision-Language models (LVLM) have spurred significant progress in document parsing task. Compared to traditional pipeline-based methods, end-to-end paradigms have shown their excellence in converting PDF images into structured outputs through integrated Optical Character Recognition (OCR), table recognition, mathematical formula recognition and so on. However, the absence of explicit analytical stages for document layouts and reading orders limits the LVLM's capability in handling complex document types such as multi-column newspapers or posters. To address this limitation, we propose in this report Logics-Parsing: an end-to-end LVLM-based model augmented with reinforcement learning. Our model incorporates meticulously designed reward mechanisms to optimize complex layout analysis and reading order inference. In addition, we expand the model's versatility by incorporating diverse data types such as chemical formulas and handwritten Chinese characters into supervised fine-tuning. Finally, to enable rigorous evaluation of our approach, we introduce LogicsParsingBench, a curated set of 1,078 page-level PDF images spanning nine major categories and over twenty sub-categories, which will be released later. Comprehensive experiments conducted on LogicsParsingBench have validated the efficacy and State-of-the-art (SOTA) performance of our proposed model across diverse document analysis scenarios. Project Page: this https URL
Abstract (translated)
最近，在大型视觉-语言模型（LVLM）方面取得的进展显著推动了文档解析任务的进步。与传统的管道式方法相比，端到端范式通过集成光学字符识别（OCR）、表格识别、数学公式识别等技术，表现出将PDF图像转换为结构化输出的强大能力。然而，缺乏明确分析文档布局和阅读顺序的阶段限制了LVLM处理如多栏报纸或海报等复杂文档类型的能力。 为了解决这一局限性，我们在本报告中提出了Logics-Parsing：一种基于LVLM的模型，并通过强化学习进行了增强。我们的模型引入了精心设计的奖励机制来优化复杂的布局分析和阅读顺序推理。此外，我们还通过监督微调将化学公式、手写中文等多样化的数据类型纳入模型中，以提高其灵活性。 为了严格评估我们的方法，我们推出了LogicsParsingBench，这是一个包含1,078张页面级别的PDF图像的数据集，涵盖了九个主要类别和二十多个子类别的文档分析场景。该数据集将在不久的将来发布。在LogicsParsingBench上进行的一系列全面实验验证了所提出模型的有效性和最先进的性能。 项目主页：[此URL](this https URL)
URL
https://arxiv.org/abs/2509.19760
PDF
-
DocIQ: A Benchmark Dataset and Feature Fusion Network for Document Image Quality Assessment
2025-09-21 10:01:43Zhichao Ma, Fan Huang, Lu Zhao, Fengjun Guo, Guangtao Zhai, Xiongkuo MinAbstract
Document image quality assessment (DIQA) is an important component for various applications, including optical character recognition (OCR), document restoration, and the evaluation of document image processing systems. In this paper, we introduce a subjective DIQA dataset DIQA-5000. The DIQA-5000 dataset comprises 5,000 document images, generated by applying multiple document enhancement techniques to 500 real-world images with diverse distortions. Each enhanced image was rated by 15 subjects across three rating dimensions: overall quality, sharpness, and color fidelity. Furthermore, we propose a specialized no-reference DIQA model that exploits document layout features to maintain quality perception at reduced resolutions to lower computational cost. Recognizing that image quality is influenced by both low-level and high-level visual features, we designed a feature fusion module to extract and integrate multi-level features from document images. To generate multi-dimensional scores, our model employs independent quality heads for each dimension to predict score distributions, allowing it to learn distinct aspects of document image quality. Experimental results demonstrate that our method outperforms current state-of-the-art general-purpose IQA models on both DIQA-5000 and an additional document image dataset focused on OCR accuracy.
Abstract (translated)
文档图像质量评估（DIQA）是包括光学字符识别（OCR）、文档恢复以及文档图像处理系统评价等应用中的一个重要组成部分。本文介绍了一个主观的DIQA数据集——DIQA-5000。该数据集中包含5,000张通过多种增强技术生成、基于500张具有不同失真的真实世界图片制作而成的文档图像。每个增强后的图像都由15位参与者在三个评价维度上进行评分：整体质量、清晰度和色彩保真度。 此外，我们提出了一种专门针对无参考DIQA模型的方法，该方法利用文档布局特征，在降低计算成本的同时保持低分辨率下的质量感知能力。考虑到图像质量受到视觉层面（包括低级和高级）特征的影响，我们设计了一个特性融合模块来从文档图像中提取并整合多层级的特征。 为了生成多维度评分，我们的模型为每个维度设定了独立的质量头部以预测评分分布，这使它能够学习到文档图像质量的不同方面。实验结果表明，在DIQA-5000数据集和一个专注于OCR准确度的额外文档图像数据集上，我们提出的方法优于当前最先进的通用IQA（Image Quality Assessment）模型。
URL
https://arxiv.org/abs/2509.17012
PDF
-
Layout-Aware OCR for Black Digital Archives with Unsupervised Evaluation
2025-09-16 16:43:34Fitsum Sileshi Beyene, Christopher L. DancyAbstract
Despite their cultural and historical significance, Black digital archives continue to be a structurally underrepresented area in AI research and infrastructure. This is especially evident in efforts to digitize historical Black newspapers, where inconsistent typography, visual degradation, and limited annotated layout data hinder accurate transcription, despite the availability of various systems that claim to handle optical character recognition (OCR) well. In this short paper, we present a layout-aware OCR pipeline tailored for Black newspaper archives and introduce an unsupervised evaluation framework suited to low-resource archival contexts. Our approach integrates synthetic layout generation, model pretraining on augmented data, and a fusion of state-of-the-art You Only Look Once (YOLO) detectors. We used three annotation-free evaluation metrics, the Semantic Coherence Score (SCS), Region Entropy (RE), and Textual Redundancy Score (TRS), which quantify linguistic fluency, informational diversity, and redundancy across OCR regions. Our evaluation on a 400-page dataset from ten Black newspaper titles demonstrates that layout-aware OCR improves structural diversity and reduces redundancy compared to full-page baselines, with modest trade-offs in coherence. Our results highlight the importance of respecting cultural layout logic in AI-driven document understanding and lay the foundation for future community-driven and ethically grounded archival AI systems.
Abstract (translated)
尽管黑人数字档案在文化和历史方面具有重要意义，但在人工智能研究和基础设施中，它们仍然处于结构性的代表性不足地位。特别是在数字化历史上黑人报纸的努力中，不一致的排版、视觉退化以及有限的标注布局数据阻碍了准确转录工作，即便存在一些系统声称能很好地处理光学字符识别（OCR）。在本文中，我们提出了一种针对黑人报纸档案的布局感知OCR流水线，并介绍了一个适合低资源存档背景的无监督评估框架。我们的方法结合了合成布局生成、增强数据上的模型预训练以及最先进的YOLO检测器融合技术。 我们使用了三个无需注释的评估指标：语义连贯性评分（SCS）、区域熵（RE）和文本冗余评分（TRS），这些指标量化了语言流畅度、信息多样性及OCR区域内冗余程度。在来自十个黑人报纸标题的400页数据集上的评估表明，布局感知OCR提升了结构多样性和减少了冗余，尽管与全页面基线相比，在连贯性方面存在一些权衡。 我们的结果突显了在人工智能驱动文档理解中尊重文化排版逻辑的重要性，并为未来基于社区和以伦理为基础的档案AI系统奠定了基础。
URL
https://arxiv.org/abs/2509.13236
PDF
-
Evaluation of Ensemble Learning Techniques for handwritten OCR Improvement
Abstract
For the bachelor project 2021 of Professor Lippert's research group, handwritten entries of historical patient records needed to be digitized using Optical Character Recognition (OCR) methods. Since the data will be used in the future, a high degree of accuracy is naturally required. Especially in the medical field this has even more importance. Ensemble Learning is a method that combines several machine learning models and is claimed to be able to achieve an increased accuracy for existing methods. For this reason, Ensemble Learning in combination with OCR is investigated in this work in order to create added value for the digitization of the patient records. It was possible to discover that ensemble learning can lead to an increased accuracy for OCR, which methods were able to achieve this and that the size of the training data set did not play a role here.
Abstract (translated)
2021年Lippert教授研究小组的学士项目中，需要使用光学字符识别（OCR）方法将历史患者的记录数字化。由于这些数据将来会被使用，因此对准确性有很高的要求，特别是在医疗领域更是如此。集成学习是一种结合多个机器学习模型的方法，并且据称能够提高现有方法的精度。为此，在本工作中探讨了集成学习与OCR相结合的可能性，以期为患者记录的数字化创造更多价值。研究发现，集成学习确实可以提高OCR的准确性，并确定了哪些方法能够实现这一点，同时发现训练数据集的大小在此过程中并不起决定性作用。
URL
https://arxiv.org/abs/2509.16221
PDF
-
Exploring Light-Weight Object Recognition for Real-Time Document Detection
2025-09-07 23:58:28Lucas Wojcik, Luiz Coelho, Roger Granada, David MenottiAbstract
Object Recognition and Document Skew Estimation have come a long way in terms of performance and efficiency. New models follow one of two directions: improving performance using larger models, and improving efficiency using smaller models. However, real-time document detection and rectification is a niche that is largely unexplored by the literature, yet it remains a vital step for automatic information retrieval from visual documents. In this work, we strive towards an efficient document detection pipeline that is satisfactory in terms of Optical Character Recognition (OCR) retrieval and faster than other available solutions. We adapt IWPOD-Net, a license plate detection network, and train it for detection on NBID, a synthetic ID card dataset. We experiment with data augmentation and cross-dataset validation with MIDV (another synthetic ID and passport document dataset) to find the optimal scenario for the model. Other methods from both the Object Recognition and Skew Estimation state-of-the-art are evaluated for comparison with our approach. We use each method to detect and rectify the document, which is then read by an OCR system. The OCR output is then evaluated using a novel OCR quality metric based on the Levenshtein distance. Since the end goal is to improve automatic information retrieval, we use the overall OCR quality as a performance metric. We observe that with a promising model, document rectification does not have to be perfect to attain state-of-the-art performance scores. We show that our model is smaller and more efficient than current state-of-the-art solutions while retaining a competitive OCR quality metric. All code is available at this https URL
Abstract (translated)
物体识别和文档倾斜估计在性能和效率方面已经有了显著的进步。新的模型主要沿着两个方向发展：一是通过使用更大的模型来提高性能，二是通过采用更小的模型来提升效率。然而，实时文档检测与校正是一个文献中尚未充分探索的小众领域，但它对于从视觉文件自动检索信息来说至关重要。在这项工作中，我们致力于开发一种高效的文档检测流水线，在光学字符识别（OCR）检索方面达到令人满意的效果，并且速度要快于现有的解决方案。 我们将IWPOD-Net（一种车牌检测网络），并对其进行了训练以在NBID（一个合成的身份证卡数据集）上进行检测任务。我们还通过使用MIDV（另一个包含合成身份证和护照文件的数据集）来进行数据增强和跨数据集验证，以此来寻找模型的最佳应用情景。我们也评估了来自物体识别和倾斜估计领域的其他顶级方法，并将其与我们的方法进行了比较。 每个方法被用来检测并校正文档，然后通过OCR系统进行读取。之后，我们使用基于Levenshtein距离的新OCR质量指标对输出结果进行评价。由于最终目标是提高自动信息检索的效率，我们将整体OCR的质量作为性能指标来衡量。我们发现即使在模型具备一定潜力的情况下，文档矫正并不需要达到完美程度就能取得顶尖的表现分数。 我们的实验结果显示，相较于当前的顶级解决方案，我们的模型不仅更小、更具效率，并且保持了相当竞争力的OCR质量指标。所有代码可在提供的网址获取。
URL
https://arxiv.org/abs/2509.06246
PDF
-
E-ARMOR: Edge case Assessment and Review of Multilingual Optical Character Recognition
2025-09-03 18:08:41Aryan Gupta, Anupam PurwarAbstract
Optical Character Recognition (OCR) in multilingual, noisy, and diverse real-world images remains a significant challenge for optical character recognition systems. With the rise of Large Vision-Language Models (LVLMs), there is growing interest in their ability to generalize and reason beyond fixed OCR pipelines. In this work, we introduce Sprinklr-Edge-OCR, a novel OCR system built specifically optimized for edge deployment in resource-constrained environments. We present a large-scale comparative evaluation of five state-of-the-art LVLMs (InternVL, Qwen, GOT OCR, LLaMA, MiniCPM) and two traditional OCR systems (Sprinklr-Edge-OCR, SuryaOCR) on a proprietary, doubly hand annotated dataset of multilingual (54 languages) images. Our benchmark covers a broad range of metrics including accuracy, semantic consistency, language coverage, computational efficiency (latency, memory, GPU usage), and deployment cost. To better reflect real-world applicability, we also conducted edge case deployment analysis, evaluating model performance on CPU only environments. Among the results, Qwen achieved the highest precision (0.54), while Sprinklr-Edge-OCR delivered the best overall F1 score (0.46) and outperformed others in efficiency, processing images 35 faster (0.17 seconds per image on average) and at less than 0.01 of the cost (0.006 USD per 1,000 images) compared to LVLM. Our findings demonstrate that the most optimal OCR systems for edge deployment are the traditional ones even in the era of LLMs due to their low compute requirements, low latency, and very high affordability.
Abstract (translated)
光学字符识别（OCR）在多语言、嘈杂且多样化的现实世界图像中仍然是一项重大挑战。随着大型视觉-语言模型（LVLM）的兴起，人们越来越关注这些模型超越固定OCR管道的能力。在此项研究中，我们介绍了一种名为Sprinklr-Edge-OCR的新颖OCR系统，该系统专门为资源受限环境中的边缘部署进行了优化。 我们在一个多语言（54种语言）、双人标注的专有数据集上对五种最先进的LVLM模型（InternVL、Qwen、GOT OCR、LLaMA和MiniCPM）以及两种传统OCR系统（Sprinklr-Edge-OCR和SuryaOCR）进行了大规模对比评估。我们的基准测试涵盖了包括准确性、语义一致性、语言覆盖率、计算效率（延迟、内存、GPU使用率）和部署成本在内的广泛指标。 为了更好地反映现实世界的适用性，我们在仅CPU环境的边缘案例部署中也进行了模型性能评估。研究结果显示，在精度方面，Qwen表现最佳（0.54），而Sprinklr-Edge-OCR在总体F1分数（0.46）上取得最优，并且在效率、处理图像速度快35倍（平均每张图片耗时0.17秒）以及成本低廉（每1,000张图片不到0.01美元，即0.006美元）方面优于其他模型。 我们的研究发现表明，在大型语言模型的时代，最适合边缘部署的OCR系统依然是传统的那些，这是由于它们对计算资源需求低、延迟短且极其经济实惠。
URL
https://arxiv.org/abs/2509.03615
PDF
-
Why Stop at Words? Unveiling the Bigger Picture through Line-Level OCR
2025-08-29 15:02:11Shashank Vempati, Nishit Anand, Gaurav Talebailkar, Arpan Garai, Chetan AroraAbstract
Conventional optical character recognition (OCR) techniques segmented each character and then recognized. This made them prone to error in character segmentation, and devoid of context to exploit language models. Advances in sequence to sequence translation in last decade led to modern techniques first detecting words and then inputting one word at a time to a model to directly output full words as sequence of characters. This allowed better utilization of language models and bypass error-prone character segmentation step. We observe that the above transition in style has moved the bottleneck in accuracy to word segmentation. Hence, in this paper, we propose a natural and logical progression from word level OCR to line-level OCR. The proposal allows to bypass errors in word detection, and provides larger sentence context for better utilization of language models. We show that the proposed technique not only improves the accuracy but also efficiency of OCR. Despite our thorough literature survey, we did not find any public dataset to train and benchmark such shift from word to line-level OCR. Hence, we also contribute a meticulously curated dataset of 251 English page images with line-level annotations. Our experimentation revealed a notable end-to-end accuracy improvement of 5.4%, underscoring the potential benefits of transitioning towards line-level OCR, especially for document images. We also report a 4 times improvement in efficiency compared to word-based pipelines. With continuous improvements in large language models, our methodology also holds potential to exploit such advances. Project Website: this https URL
Abstract (translated)
传统的光学字符识别（OCR）技术是先对每个字符进行分割，然后逐一识别。这种方法在字符分割阶段容易出错，并且缺乏上下文信息来利用语言模型。近年来，在序列到序列翻译领域的进展促使现代技术首先检测单词，然后再一次输入一个单词至模型中以直接输出完整的单词字符序列。这使得更好地利用语言模型成为可能，并跳过了易出错的字符分割步骤。 我们观察到上述风格的变化将准确性瓶颈转移到了词级切分上。因此，在本文中，我们提出了从词级OCR自然且逻辑地过渡到行级OCR的方法。该方法能够规避单词检测中的错误，并提供更大的句子上下文以更好地利用语言模型。我们表明所提出的这种方法不仅提高了OCR的准确率，还提升了其效率。 尽管进行了详尽的文献调查，但我们没有找到任何可用于训练和评估从词级转向行级OCR转变的公开数据集。因此，我们也贡献了一个精心策划的数据集，包含251张英文页面图像，并附有行级别注释。我们的实验结果显示了显著的整体准确率提升达5.4%，突显了向行级OCR转换的潜在好处，特别是对文档图像而言。此外，相比基于单词的流水线方法，我们还报告了效率提高了四倍。 随着大型语言模型的持续改进，我们的方法也具有利用这些进步的潜力。项目网站：[此链接](https://example.com/)（请将示例网址替换为实际提供的项目网址）
URL
https://arxiv.org/abs/2508.21693
PDF
-
Exploration of Deep Learning Based Recognition for Urdu Text
2025-08-18 10:27:48Sumaiya Fazal, Sheeraz AhmedAbstract
Urdu is a cursive script language and has similarities with Arabic and many other South Asian languages. Urdu is difficult to classify due to its complex geometrical and morphological structure. Character classification can be processed further if segmentation technique is efficient, but due to context sensitivity in Urdu, segmentation-based recognition often results with high error rate. Our proposed approach for Urdu optical character recognition system is a component-based classification relying on automatic feature learning technique called convolutional neural network. CNN is trained and tested on Urdu text dataset, which is generated through permutation process of three characters and further proceeds to discarding unnecessary images by applying connected component technique in order to obtain ligature only. Hierarchical neural network is implemented with two levels to deal with three degrees of character permutations and component classification Our model successfully achieved 0.99% for component classification.
Abstract (translated)
乌尔都语是一种连笔书写语言，与阿拉伯语及其他许多南亚语言有相似之处。由于其复杂的几何结构和形态特征，乌尔都语难以分类。如果分割技术高效，则可以进一步处理字符分类，但由于乌尔都语的上下文敏感性，基于分割的方法通常会导致较高的错误率。我们提出的乌尔都语光学字符识别系统方法是一种基于组件的分类法，依赖于自动特征学习技术——卷积神经网络（CNN）。该 CNN 在通过三个字符排列生成的乌尔都语文本数据集上进行训练和测试，并进一步采用连通成分技术丢弃不必要的图像以获取连字。我们实施了具有两个级别的分层神经网络来处理三种字符排列程度及组件分类。我们的模型在组件分类方面成功达到了 0.99% 的精度。
URL
https://arxiv.org/abs/2508.13245
PDF
