Video text detection and localization based on localized generalization error model

Ma, Xianheng; Ng, Wing W. Y.; Chan, Patrick P. K.; Yeung, Daniel

doi:10.1109/icmlc.2010.5580484

Cited by 5 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kim et al [33] proposed a texture-based method using support vector machines (SVM) to classify text and nontext pixels, the method employs an adaptive mean shift algorithm along with SVM.Edge and texture features without classifier were proposed by Liu and Dai in [23] for text detection, but the method uses a large number of features to discriminate text and nontext pixels. Edge-texture features with a classifier are also used for video text detection [24], [25]. But this method is sensitive to complex background for accurate text detection in videos.Although the method works well for a variety of frames, it requires more time to process due to the large number of features and wavelet transforms.The combination of edge based features and texture based features used for efficient text detection are carried but there is a problems to produce few false positive text [26].…”

Section: Iirelated Workmentioning

confidence: 99%

Video frames text detection through Bayesian classification and boundary growing method

Nancy¹,

Jayapriya²

2014

International Conference on Information Communication and Embedded Systems (ICICES2014)

View full text Add to dashboard Cite

Text detection in natural scene images is important for content-based image analysis. It consists of different component analysis for text detection video frames.Multioriented text detection in video frames is one of the text detection in videos frames. Multi -oriented text detection is not easy as detection of captions or graphics which is usually appears in horizontal direction and has high contrast compared to its background. Multi-oriented text detection generally refers to scene text that makes text detection more challenging. Therefore conventional text detection may not give good results for multioriented video frames text detection.Hence, in this paper, we present a new enhancement method that includes the product of Laplacian and Sobel operations to enhance text pixels in videos. To classify true text pixels, we propose a Bayesian classifier without assuming a priori probability about the input frame but estimating it based on three probable matrices. Three different ways of clustering are performed on the output of the enhancement method to obtain the three probable matrices. Text candidates are obtained by intersecting the output of the Bayesian classifier with the Canny edge map of the input frame. The robustness of the methods are analyzed our own datasets under different video frames. Finally, it coarse-to-fine detection locates text regions efficiently.keywords-Bayesian classifier, boundary growing, Laplacian-Sobel product (LSP), maximum gradient difference, multioriented video scene text detection, text candidate detection

show abstract

Section: Iirelated Workmentioning

confidence: 99%