Scene text detection and recognition: a survey

Naiemi, Fatemeh; Ghods, Vahid; Khalesi, Hassan

doi:10.1007/s11042-022-12693-7

Cited by 27 publications

(12 citation statements)

References 97 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Increment in the standard size of the empirical data, along with a considerable increase in the computational complexity of various processing tasks, from basic processes like image correction, compulsory registration, essential quality, and contrast enhancement, and filtering to sophisticated processes such as the alleged discrimination of real texts from fake ones incorrectly are very critical tasks. Moreover, the other particular tasks are the determination of the font size, discrimination of the overlapped texts, correction of the text rotation angle, and correction of the convexity and concavity of the background image, which necessitates applying strong processing structures with fast parallel architectures and specific distributed algorithms for big image data processing, in addition to benefiting from parallel and efficient distributed software methods [59].…”

Section: Case Studymentioning

confidence: 99%

“…Text localization for automatic applications such as sign and driving traffic sign recognition, reading old texts, archiving documents, finding news feeds, and recognition of the text on different locations like emergency vehicles, clothes, and billboards are fully developed with the aid of static and flying/portable robots. The classic texts in typical scenes and landscapes are more diverse and unpredictable in direct comparison to structured graphical texts [59,61]. The successful extraction of classic texts from these possible kinds of emotional scenes typically using portable robots for accurate navigation, graciously according to traffic signs, the direct detection and critical recognition of license plates, object recognition, and so on are examples of novel applications of automatic text recognition [62].…”

Section: Case Studymentioning

confidence: 99%

See 1 more Smart Citation

Big Image Data 15Vs Model for Intelligent Data Ocean of Multimedia Things

Mahabadi,

Abdolazimi

2024

Preprint

View full text Add to dashboard Cite

The classification of large image data using traditional methods has been based on changes in image gray features, extraction of edge and contour feature information, or conversion between image coordinate sets. However, due to the growth of new image size and big image data (BID) in real-time multimedia communication and future online big data (BD) applications, these methods have become increasingly complex and resulting in poor real-time performance. These complex methods suffer from complex algorithms, massive data communication, slow processing speed, unintelligent predictive modeling, weak data classification, limited accuracy, uncleaned usage data, combined destructive artificial and natural noise, exchanging data values over time, and exhausting updating data for accurate operation of media storage. In the era of the Internet of Multimedia Things (IoMT), most modern devices can accurately capture vast amounts of observational data in the form of valuable images, text, and acoustic recordings. This reliable data must be kept valid for local data processing at different times and for globally extracting knowledge from information at various data levels to address forthcoming challenges related to big image data. The challenges of secure communication require accurate, reliable, fast, and effective information gathering for local decision-making and global knowledge mining of BID to support cyber-physical systems for speedy data virtualization of smart city devices. Complex and uncontrollable problems persist on the outskirts of grids, fogs, and cloud networks regarding data cleansing and privacy protection for intelligent data ocean management. This study developed a new 15Vs model empirically to examine distributed big image data processing based on a modern BID connectivity approach. The model accurately extracts textured features from visible data images and overcomes a unique set of key challenges. The model introduces a strategic analyzer, uses intelligent local agents, and recruits clever global bots to provide a suitable platform for generous support of bot-oriented BD processing. This instantly forms a hierarchical data level for better data acquisition and cleansing, safe privacy protection, suitable information diffusion, and speedy knowledge extraction. Our study presents a novel definition of BID to address the ultimate challenges of data management in a high-risk environment for further big image data modeling research needs.

show abstract

Section: Case Studymentioning

confidence: 99%

Section: Case Studymentioning

confidence: 99%

Big Image Data 15Vs Model for Intelligent Data Ocean of Multimedia Things

Mahabadi,

Abdolazimi

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Many algorithms are greatly influenced by and modelled after object detectors. In general, the development of scene text detection algorithms is divided into three stages [27]. Learning-based techniques are equipped with multi-step pipelines in the first phase to replace handcrafted features.…”

Section: ) Text Detectionmentioning

confidence: 99%

SignboardText: Text Detection and Recognition in In-the-Wild Signboard Images

do,

Tran,

Nguyen

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Scene text detection and recognition have attracted much attention in recent years because of their potential applications. Detecting and recognizing texts in images may suffer from scene complexity and text variations. Some of these problematic cases are included in popular benchmark datasets, but only to a limited extent. In this work, we investigate the problem of scene text detection and recognition in a domain with extreme challenges. We focus on in-thewild signboard images in which text commonly appears in different fonts, sizes, artistic styles, or languages with cluttered backgrounds. We first contribute an in-the-wild signboard dataset with 79K text instances on both line-level and word-level across 2,104 scene images. We then comprehensively evaluated recent state-of-the-art (SOTA) approaches for text detection and recognition on the dataset. By doing this, we expect to realize the barriers of current stateof-the-art approaches to solving the extremely challenging issues of scene text detection and recognition, as well as their applicability in this domain. Code and dataset are available at https://github.com/aiclub-uit/SignboardText/ and IEEE DataPort.

show abstract

“…Literatürde sabit ve hareketli görüntü dosyalarından altyazıların elde edilmesi ile ilgili pek çok çalışma yapılmıştır. Bu çalışmaların temel amacı görüntü içerisinde metnin varlığının belirlenmesi [4,6,7], yerelleştirilmesi [8][9][10][11][12] ve tanınması [6,13] işlemlerini kapsamaktadır. Görüntü çerçevelerinde yer alan metinlerin elde edilmesi amacıyla bir çalışma [14] tarafından yapılmıştır.…”

Section: Introductionunclassified

Detecting subtitle regions in multimedia images using image processing techniques

GÜVENOĞLU

2023

Artıbilim: Adana Alparslan Türkeş Bilim Ve Teknoloji Üniversitesi Fen Bilimleri Dergisi

View full text Add to dashboard Cite

Mobil cihazların ve multimedya teknolojilerinin yaygın olarak kullanımı ile birlikte görüntülerin elde edilmesi çok daha kolay hale gelmiştir. Bununla birlikte görüntüler içerisinde yer alan alt yazıların elde edilmesi ve bunların farklı amaçlar için kullanımı bir problem olarak ortaya çıkmıştır. Bu çalışmada multimedya görüntülerinde yer alan altyazıların bulunduğu bölgelerin tespit edilmesi için kullanımı basit ve etkili bir yöntem önerilmiştir. Yöntem birbirini takip eden farklı adımlardan oluşmaktadır. Giriş görüntüsü olarak verilen 24 bit renkli bir görüntüler üzerinde görüntü işleme teknikleri yardımıyla metin olan alt yazı bölgesine ait koordinatlar belirlenmektedir. Ardından renkli görüntü üzerinde işaretlenmektedir. Birbirinden farklı özellik ve ölçülerde görüntüler üzerinde deneysel çalışmalar gerçekleştirilmiştir. Çalışmanın gerçekleştirilmesinde köşe noktaların işaretlenmesi amacıyla Harris köşe saptama algoritması, gürültülerin giderilmesi için gauss filtreleme ve morfolojik görüntü işleme teknikleri kullanılmıştır. Gerçekleştirilen çalışmalarda %94 oranında bir başarım elde edilmiştir. Süre ölçüm testlerinde ise ortalama olarak 1.56 sn gibi iyi bir başarım süresine ulaşılmıştır. Süre ölçümleri literatürdeki diğer çalışmalar ile karşılaştırılmıştır. Önerilen yöntemin, süre bakımından oldukça iyi bir performansa sahip olduğu görülmüştür.

show abstract

Scene text detection and recognition: a survey

Cited by 27 publications

References 97 publications

Big Image Data 15Vs Model for Intelligent Data Ocean of Multimedia Things

Big Image Data 15Vs Model for Intelligent Data Ocean of Multimedia Things

SignboardText: Text Detection and Recognition in In-the-Wild Signboard Images

Detecting subtitle regions in multimedia images using image processing techniques

Contact Info

Product

Resources

About