A robust arbitrary text detection system for natural scene images

Risnumawan, Anhar; Shivakumara, Palaiahnakote; Chan, Chee Seng; Tan, Chew Lim

doi:10.1016/j.eswa.2014.07.008

Cited by 448 publications

(183 citation statements)

References 29 publications

Supporting

Mentioning

180

Contrasting

Unclassified

Order By: Relevance

“…This process continues until no text candidates exist in the seed cluster. There are other approaches [7,26,33] that explore the growing process in different ways to extract arbitrarily oriented text lines, but these approaches do not use the symmetry concept. Thus the main advantage of this process is that it works for any direction of text line.…”

Section: Symmetry Driven Nearest Neighbor For Arbitrary Text Line mentioning

confidence: 99%

Multi-Spectral Fusion Based Approach for Arbitrarily Oriented Scene Text Detection in Video Images

Liang

Shivakumara

Lü

et al. 2015

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

Scene text detection from video as well as natural scene images is challenging due to the variations in background, contrast, text type, font type, font size, and so on. Besides, arbitrary orientations of texts with multi-scripts add more complexity to the problem. The proposed approach introduces a new idea of convolving Laplacian with wavelet sub-bands at different levels in the frequency domain for enhancing low resolution text pixels. Then, the results obtained from different sub-bands (spectral) are fused for detecting candidate text pixels. We explore maxima stable extreme regions along with stroke width transform for detecting candidate text regions. Text alignment is done based on the distance between the nearest neighbor clusters of candidate text regions. In addition, the approach presents a new symmetry driven nearest neighbor for restoring full text lines. We conduct experiments on our collected video data as well as several benchmark data sets, such as ICDAR 2011, ICDAR 2013, and MSRA-TD500 to evaluate the proposed method. The proposed approach is compared with the state-of-the-art methods to show its superiority to the existing methods.

show abstract

Section: Symmetry Driven Nearest Neighbor For Arbitrary Text Line mentioning

confidence: 99%

Multi-Spectral Fusion Based Approach for Arbitrarily Oriented Scene Text Detection in Video Images

Liang

Shivakumara

Lü

et al. 2015

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Thus, two major issues arise for natural scene text deblurring: the text detection and the modeling of text properties. To accomplish this, we first need using some text localization methods [14], [32], [33] to detect text areas. In this section, the text field is detected by [14], based on which we learn the TMD for the text fields, and a natural dictionary for the non-text one.…”

Section: The Proposed Approach Based On Tmdmentioning

confidence: 99%

“…6. Besides, we could use more effective and sophisticated localization methods, such as [32] and [33], to obtain more accurate results. We illustrate the proposed algorithm with a simple example in Fig.…”

Section: A An Illustrative Examplementioning

confidence: 99%

Scene Text Deblurring Using Text-Specific Multiscale Dictionaries

Cao

Ren

Zuo

et al. 2015

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Texts in natural scenes carry critical semantic clues for understanding images. When capturing natural scene images, especially by handheld cameras, a common artifact, i.e., blur, frequently happens. To improve the visual quality of such images, deblurring techniques are desired, which also play an important role in character recognition and image understanding. In this paper, we study the problem of recovering the clear scene text by exploiting the text field characteristics. A series of text-specific multiscale dictionaries (TMD) and a natural scene dictionary is learned for separately modeling the priors on the text and nontext fields. The TMD-based text field reconstruction helps to deal with the different scales of strings in a blurry image effectively. Furthermore, an adaptive version of nonuniform deblurring method is proposed to efficiently solve the real-world spatially varying problem. Dictionary learning allows more flexible modeling with respect to the text field property, and the combination with the nonuniform method is more appropriate in real situations where blur kernel sizes are depth dependent. Experimental results show that the proposed method achieves the deblurring results with better visual quality than the state-of-the-art methods.

show abstract

“…However, these methods are developed for English text detection and recognition but not for other textssuch as Indus texts. It is noted from the methods ( [10]; [20]; [24]; [25]) that achieving high recognition rate for degraded and blurred text is still considered as an open issue in the field of video and camera-based image analysis. Therefore, we can conclude that these methods require individual scripts to achieve a high success rate for recognition.…”

Section: 0introductionmentioning

confidence: 99%

“…In addition, there is no universal OCR available for recognizing multiple scripts. There are also methods which address the problem of poor quality text in low contrast images ( [10]; [20]; [24]; [25]) for detection and recognition. However, these methods are developed for English text detection and recognition but not for other textssuch as Indus texts.…”

Section: 0introductionmentioning

confidence: 99%