An Overview of the Tesseract OCR Engine

By Raymond W. Smith. 2007

2,690 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Towards VQA Models That Can Read (2019, methodology): “We assume that the OCR model can read and return word tokens from an image, e.g. [6, 41].”
TextCaps: a Dataset for Image Captioning with Reading Comprehension (2020, background): “In recent years, with the availability of large labelled corpora, progress in image captioning has seen steady increase in performance and quality [4,10,12,13,37] and reading scene text (OCR) has matured [8,16,22,24,34].”
Approximate Bayesian Image Interpretation using Generative Probabilistic Graphics Programs (2013, methodology): “For example, in our text application, the Sis include binary indicators for the presence or absence of each glyph, along with its identity (“A“ through ”Z”, plus digits 0-9), and parameters including location,”
Text String Detection from Natural Scenes by Structure-based Partition and Grouping (2011, background): “, most part of the captured image contains well organized text with clean background), many algorithms and commercial optical character recognition (OCR) systems have been developed [2], [32].”
Multimodal Attention with Image Text Spatial Relationship for OCR-Based Image Captioning (2020, background): “Inspired by the nearly mature technologies for Optical Character Recognition (OCR) [8, 12, 19, 23, 32], a pioneering work first attempts to include image text in generated descriptions and proposes the task of image captioning based on reading and understanding image text [30].”
Extracting code from programming tutorial videos (2016, methodology): “ACE is implemented both in Java and Python and relies on OpenCV [3] for frame extraction and image processing, on the Tesseract OCR library [18] for text extraction, on scikit-learn [14] for clustering, and on ANTLR [13] for parsing and lexing.”
PlotQA: Reasoning over Scientific Plots (2019, methodology): “For extracting this data from within these bounding boxes, we use a state-of-the-art OCR model [34].”
NL-Augmenter: A Framework for Task-Sensitive Natural Language Augmentation (2021, methodology): “It renders the input sentence as an image and recognizes the rendered text using the OCR engine Tesseract 4 (Smith, 2007).”
FUNSD: A Dataset for Form Understanding in Noisy Scanned Documents (2019, methodology): “We evaluate two OCR engines for text recognition: Tesseract [15] and Google Vision.”
Sensor-Based Assistive Devices for Visually-Impaired People: Current Status, Challenges, and Future Directions (2017, methodology): “An assistive navigator integrated both range and visual information Navigation Assistance Using RGB-D Sensor with Range Expansion (Nav RGB-D) An assistive navigator integrated both range and visual This proposed device can be more than a navigator for blind people; it can be a light flash for…”
High-Performance OCR for Printed English and Fraktur Using LSTM Networks (2013, methodology): “Instead, we tested performance on two books that we use as test cases: Theodor Fontane Wanderungen durch die Mark Brandenburg (a clean, high resolution scan) TABLE I: Character error rates of the LSTM recognizer (without a language model or adaptation), the Tesseract [2], ABBYY commercial OCR system”
Using text-spotting to query the world (2010, methodology): “We evaluated two open-source OCR packages: GOCR and Tesseract [23] and chose the latter.”
Outside the Closed World: On Using Machine Learning for Network Intrusion Detection (2010, background): “Examples from other domains include product recommendations systems such as used by Amazon [3] and Netﬂix [4]; optical character recognition systems (e.g., [5], [6]); natural language translation [7]; and also spam detection, as an example closer to home [8].”
Improving patch-based scene text script identification with ensembles of conjoined networks (2016, methodology): “Figure 8 shows the confusion matrices for our method in all three datasets with detailed per class classification results.”
Finding and Characterizing Information Warfare Campaigns (2020, methodology): “OCR on preprocessed images was accomplished with Google Tesseract [219].”
Invoice Haystack: Benchmarking Document Retrieval and Visual Question Answering Under Strong Visual Homogeneity (2026, unknown): “Document AI has evolved from modular pipelines combining OCR [3,62], layout analysis [59, 77], and NLP models [18,45] toward unified end-to-end architectures.”
Leveraging Vision-Language Models for Visual Grounding and Analysis of Automotive UI (2025, unknown): “To establish a suitable baseline model for our fine-tuning and experiments, we evaluated four existing VLMs: TinyClick (0.27B) [24], Molmo-7B-D-0924 [4], UGround-V1-7B (Qwen2-VL-based) [10], and InternVL2 5-8B [2].”