Efficient video text detection using edge features

Palaiahnakote, Shivakumara; Huang, Wei‐Hua; Tan, Chew Lim

doi:10.1109/icpr.2008.4761415

Cited by 46 publications

(24 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second attractive feature is that it processes oriented text as the text orientations and are approximated in the localization step. Language independent features or multilingual OCR modules in [12], [27], [34], processes multilingual text. The disadvantages are dual.…”

Section: Methodologiesmentioning

confidence: 99%

“…A review of the new advancements achieved in the text recognition in multimedia documents is presented in [36]. Various issues related to the recognition problem have been identified, including text detection [9,21,23,46], Enhancement of Text and linearization [3,15,24,20, 48] (a pre-processing step that aims to improve recognition performance), character segmentation [2,22,31,34], and its recognition [4,17,32] and the integration of linguistic knowledge [11,27,49].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Automated Recognition of Text in Images: A Survey

Wadhawan¹,

Gajendran²

2015

IJCA

View full text Add to dashboard Cite

OCR (Optical Character Recognition) System works in the domain of Natural Language Processing and Image Processing. This is used to convert all the text information that is present in image form, to text format. Text is one of the most influential inventions of Humanity. The fertile and precise information incorporated in text is very useful in a wide range of applications that are computer-vision based, and hence text detection and recognition in natural scenes (e.g.: traffic sign boards, license plate, Hoardings and videos etc.) have become important and active research topics in computer vision and document analysis. This survey paper presents a review of various state-of-the-art techniques proposed for different processes (i.e. detection, localization, extraction, etc.) of text information processing in Images. Literature review can further serve as a good reference for researchers in the areas of scene text detection and recognition. The aim is to introduce the researchers to the latest trends in this area and to serve as a resource for developers who wish to integrate such solutions into their own work.

show abstract

Section: Methodologiesmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Automated Recognition of Text in Images: A Survey

Wadhawan¹,

Gajendran²

2015

IJCA

View full text Add to dashboard Cite

show abstract

“…For example, Li and Wang used thresholds on the width and height of text blocks, along with other metrics such as the ratio between edge area and text area [4]. Shivakumara et al used profile projections, edge density, and character distances to filter out false positive regions [5][6][7]. Kim and Kim used aspect ratios to verify the proposed text regions [8].…”

Section: Text Verificationmentioning

confidence: 99%

An Image Dataset of Text Patches in Everyday Scenes

Ahmed

Abbott

Hussein

2016

Advances in Visual Computing

View full text Add to dashboard Cite

Abstract. This paper describes a dataset containing small images of text from everyday scenes. The purpose of the dataset is to support the development of new automated systems that can detect and analyze text. Although much research has been devoted to text detection and recognition in scanned documents, relatively little attention has been given to text detection in other types of images, such as photographs that are posted on social-media sites. This new dataset, known as COCO-Text-Patch, contains approximately 354,000 small images that are each labeled as "text" or "non-text". This dataset particularly addresses the problem of text verification, which is an essential stage in the end-to-end text detection and recognition pipeline. In order to evaluate the utility of this dataset, it has been used to train two deep convolution neural networks to distinguish text from non-text. One network is inspired by the GoogLeNet architecture, and the second one is based on CaffeNet. Accuracy levels of 90.2% and 90.9% were obtained using the two networks, respectively. All of the images, source code, and deep-learning trained models described in this paper will be publicly available 5 .

show abstract

“…where p x,y is a pixel of x, y coordinates, L is the number of pixels in the local variance mask and I vertical t , I horizontal t , I diagonal t are vertical, horizontal, and diagonal gradients of a frame t. These kind of gradients are proven to work well in text detection process [30].…”

Section: I = I I Tt−τ ∨ I I Tt−2τ ∨ I I T−τt−2τmentioning

confidence: 99%

Real-Time Event Detection in Field Sport Videos

Kapela¹,

McGuinness

Świetlicka³

et al. 2014

Computer Vision in Sports

View full text Add to dashboard Cite

This chapter describes a real-time system for event detection in sports broadcasts. The approach presented is applicable to a wide range of field sports. Using two independent event detection approaches that work simultaneously, the system is capable of accurately detecting scores, near misses, and other exciting parts of a game that do not result in a score. The results obtained across a diverse dataset of different field sports are promising, demonstrating over 90 % accuracy for a feature-based event detector and 100 % accuracy for a scoreboard-based detector detecting only scores. Event Detection in SportsSport has always been one of the most popular of television broadcasts [1,2]. Sports broadcast viewing figures are consistently high, particularly for events like the Olympics and national or regional finals of nationally popular games. Given such wide appeal and popularity, there has been significant interest in algorithms for automatic event detection in sports broadcasts [3]. This is motivated by potential applications such as automatic highlight generation for summarization for emerging second screen applications, indexing for search, and retrieval in large archives, mobile content delivery either offline or as an added value in-stadium user experience.

show abstract

Efficient video text detection using edge features

Cited by 46 publications

References 8 publications

Automated Recognition of Text in Images: A Survey

Automated Recognition of Text in Images: A Survey

An Image Dataset of Text Patches in Everyday Scenes

Real-Time Event Detection in Field Sport Videos

Contact Info

Product

Resources

About