Scene Text Detection and Recognition Using Deep Learning
A special issue of Sensors (ISSN 1424-8220). This special issue belongs to the section "Sensing and Imaging".
Deadline for manuscript submissions: 10 December 2025
Special Issue Editors
Interests: information engineering; spectroscopy; spectral imaging; plant phenotyping; agro-food quality detection; machine learning
Special Issues, Collections and Topics in MDPI journals
Interests: Cramer-Rao Lower Bound,Semidefinite Programming,Semidefinite Programming Problem,Convex Form,Covariance Matrix,Global Convergence,Local Convergence,Performance Of Solutions,Position Estimation,Root Mean Square Error,Semidefinite Relaxation,Time Difference
Special Issue Information
Scene text detection and recognition are central in various real-world applications, including intelligent transportation systems, smart cities, augmented reality, and social media content analysis. In modern urban environments, text appears in a variety of forms—on traffic signs, shop boards, advertisements, and digital media—often under complex conditions such as diverse fonts, arbitrary orientations, occlusions, and varying text sizes. Detecting and recognizing text from these natural scenes is a key challenge in computer vision.
This topic focuses on scene text detection and recognition across various languages rather than limiting the focus to any specific language. In real-world scenes, text can appear in a combination of languages (e.g., Chinese, English, Arabic, etc.), and each language has its unique script, layout, and context. The challenge is developing models that can handle text detection and recognition for a wide range of scripts in languages such as Latin, Chinese, Arabic, etc. Large language models (LLMs), such as GPT, LLaMA, and vision-language models (e.g., CLIP, GPT-4V, and LLaVA), have shown great potential in improving scene text understanding by providing text recognition, correction, and semantic interpretation capabilities across multiple languages.
Alongside modern text recognition, ancient Chinese scene text detection and recognition represent a unique research domain. Ancient Chinese texts—found in stone inscriptions, murals, scrolls, and other historical artifacts—pose significant challenges due to their diverse calligraphic forms (seal, clerical, and cursive scripts), damage, erosion, and the historical context in which they were written. Detecting and recognizing these texts requires not only computer vision techniques but also deep knowledge of linguistics, archeology, and cultural heritage preservation.
This issue encourages research that advances state-of-the-art scene text detection across multiple languages and ancient Chinese text recognition, emphasizing the importance of leveraging large language models and multimodal learning frameworks.
Topics of interest include, but are not limited to, the following:
- Scene text detection and recognition for various languages (e.g., Chinese, English, Arabic, and others);
- End-to-end models integrating text detection and recognition for multi-lingual or multi-script environments;
- Large language model-assisted recognition, correction, and contextual understanding for multi-language text;
- The detection and recognition of ancient Chinese scripts (seal, clerical, cursive) on historical artifacts;
- Character enhancement and denoising techniques for damaged, eroded, or low-contrast text;
- Domain adaptation from modern text to ancient Chinese or other historical scripts;
- Few-shot and zero-shot learning for rare characters and cross-language applications;
- Vision–language models for scene text recognition and multi-lingual semantic understanding (e.g., CLIP, GPT-4V, and LLaVA);
- Robustness in scene text detection under occlusion, low-resolution, blur, and noisy backgrounds;
- Applications in smart signage, public safety, digital humanities, and cultural heritage preservation.
We encourage the submission of contributions from researchers working on multi-language scene text detection and recognition, including those advancing techniques for ancient Chinese, with a focus on utilizing modern AI methods and large language models.
Dr. Chu Zhang
Prof. Dr. Hengnian Qi
Guest Editors
Manuscript Submission Information
Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.
Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.
Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.
Keywords
- Scene text detection
- Scene text recognition
- Multi-lingual text recognition
- Ancient Chinese text recognition
- Large language models (LLMs)
- Vision–language models
- Few-shot learning
- Zero-shot learning
- Document layout understanding
- Cultural heritage preservation.
Benefits of Publishing in a Special Issue
- Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
- Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
- Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
- External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
- Reprint: MDPI Books provides the opportunity to republish successful Special Issues in book format, both online and in print.
Further information on MDPI's Special Issue policies can be found here.
