Filters








14,414 Hits in 5.3 sec

Text detection from scene images using sparse representation

Wumo Pan, T. D. Bui, C. Y. Suen
2008 Pattern Recognition (ICPR), Proceedings of the International Conference on  
A sparse representation based method is proposed for text detection from scene images.  ...  The core of the labeling process is a sparsity test using an over-complete dictionary, which is learned from edge segments of isolated character images.  ...  Introduction Text detection and recognition from scene images is a very challenging problem and is attracting more and more research effort from the document image analysis community.  ... 
doi:10.1109/icpr.2008.4760967 dblp:conf/icpr/PanBS08 fatcat:epbgtfcpn5fztlb7546phovmli

Restoration of Degraded Images for Text Detection and Recognition

Sayali R., Sankirti S.
2016 International Journal of Computer Applications  
This research focuses on removing a maximum number of degradation factors from a natural scene image containing text such that the detection and recognition of the text present in that image becomes very  ...  The task of text detection natural scene images is very challenging due to the complex background and unpredictable text appearances in the image.  ...  This research presents a novel idea of restoring text in natural scene images by using dictionaries of sparse representations.  ... 
doi:10.5120/ijca2016907895 fatcat:zuyllbyokva3nmjmxbobd3lrxq

Robust Scene Text Recognition Using Sparse Coding based Features [article]

Da-Han Wang, Hanzi Wang, Dong Zhang, Jonathan Li, David Zhang
2015 arXiv   pre-print
In this paper, we propose an effective scene text recognition method using sparse coding based features, called Histograms of Sparse Codes (HSC) features.  ...  The HSC features are extracted by computing sparse codes with dictionaries that are learned from data using K-SVD, and aggregating per-pixel sparse codes to form local histograms.  ...  First, we propose an effective scene text recognition method using sparse coding based features (i.e., the HSC features) that are learned automatically from data for character feature representation.  ... 
arXiv:1512.08669v1 fatcat:yivuztc3lbbavitkez474d2x44

Text detection in images using sparse representation with discriminative dictionaries

Ming Zhao, Shutao Li, James Kwok
2010 Image and Vision Computing  
In this paper, we propose a classification-based algorithm for text detection using a sparse representation with discriminative dictionaries.  ...  Extensive experiments show that the proposed method can effectively detect texts of various sizes, fonts and colors from images and videos.  ...  [17] proposed the use of a sparse representation for text detection. It extracts text-like edges from an image by using a dictionary obtained by K-SVD [20] .  ... 
doi:10.1016/j.imavis.2010.04.002 fatcat:z7ndqrbqpjejvo2ysogcnt6zau

Natural Scene Character Recognition Using Robust PCA and Sparse Representation

Zheng Zhang, Yong Xu, Cheng-Lin Liu
2016 2016 12th IAPR Workshop on Document Analysis Systems (DAS)  
Histogram of oriented Gradient (HOG) to perform image feature extraction, and finally, use a sparse representation based classifier for recognition.  ...  Natural scene character recognition is challenging due to the cluttered background, which is hard to separate from text.  ...  These conventional OCR methods perform satisfactorily on scanned document images, but for extracting text from natural scene images, they cannot work well because the separation of text from image background  ... 
doi:10.1109/das.2016.32 dblp:conf/das/ZhangXL16 fatcat:5pjjknya5vhbvgwbyoo2gvymh4

Scene Text Deblurring Using Text-Specific Multiscale Dictionaries

Xiaochun Cao, Wenqi Ren, Wangmeng Zuo, Xiaojie Guo, Hassan Foroosh
2015 IEEE Transactions on Image Processing  
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.  ...  In this paper, we study the problem of recovering the clear scene text by exploiting the text field characteristics.  ...  Therefore, we design the prior of scene text based on sparse representation in this paper for scene text image deblurring.  ... 
doi:10.1109/tip.2015.2400217 pmid:25705915 fatcat:e3ziarcjw5dgtbufwmizzzf2nq

Feature Representations for Scene Text Character Recognition: A Comparative Study

Chucai Yi, Xiaodong Yang, Yingli Tian
2013 2013 12th International Conference on Document Analysis and Recognition  
Recognizing text character from natural scene images is a challenging problem due to background interferences and multiple character patterns.  ...  Scene Text Character (STC) recognition, which generally includes feature representation to model character structure and multi-class classification to predict label and score of character class, mostly  ...  ICDAR2003 dataset is prepared for the Robust Reading Competition of scene text detection and recognition. It has 509 scene images and 2268 word-level text regions.  ... 
doi:10.1109/icdar.2013.185 dblp:conf/icdar/YiYT13 fatcat:euxmmkqdpfbm5avt6cqknzgfgy

Video text localization using wavelet and shearlet transforms

Purnendu Banerjee, B. B. Chaudhuri, Bertrand Coüasnon, Eric K. Ringger
2013 Document Recognition and Retrieval XXI  
In this paper, we present a new method of text detection using a combined dictionary consisting of wavelets and a recently introduced transform called shearlets.  ...  Then K-means clustering is used for obtaining text pixels from the Standard Deviation (SD) of combined coefficient of wavelets and shearlets as well as the union of wavelets and shearlets features.  ...  Scene text detection result of several test images from ICDAR 2011 competition dataset are given in Figure 6 .  ... 
doi:10.1117/12.2036077 dblp:conf/drr/BanerjeeC14 fatcat:z6qlid36wbefhdeslphxag3ksa

Learning Co-occurrence of Local Spatial Strokes for Robust Character Recognition

Song GAO, Chunheng WANG, Baihua XIAO, Cunzhao SHI, Wen ZHOU, Zhong ZHANG
2014 IEICE transactions on information and systems  
In this paper, we propose a representation method based on local spatial strokes for scene character recognition.  ...  The encouraging results outperform state-of-the-art algorithms. key words: robust character recognition, local spatial stroke, cooccurrence, sparse dictionary  ...  Text detection and text recognition are performed sequentially to extract scene text information. In the text detection stage, regions containing scene texts are localized from entire images.  ... 
doi:10.1587/transinf.e97.d.1937 fatcat:xqg6ihrkifb3fhfts46tifbf44

Multilingual Text Detection with Nonlinear Neural Network

Lin Li, Shengsheng Yu, Luo Zhong, Xiaozhen Li
2015 Mathematical Problems in Engineering  
Multilingual text detection in natural scenes is still a challenging task in computer vision.  ...  We also develop a novel nonlinear network based on traditional Convolutional Neural Network that is able to detect multilingual text regions in the images.  ...  These images are taken from indoor (office and mall) and outdoor (street) scenes using a packet camera. KAIST provides a scene text dataset consisting of 3000 images of indoor and outdoor scenes.  ... 
doi:10.1155/2015/431608 fatcat:gsw3rhy755cphjl3wwk2v5on5u

Linear Spatial Pyramid Matching Using Non-convex and non-negative Sparse Coding for Image Classification [article]

Chengqiang Bao and Liangtian He and Yilun Wang
2015 arXiv   pre-print
Recently sparse coding have been highly successful in image classification mainly due to its capability of incorporating the sparsity of image representation.  ...  Our numerical experiments show that the improved approach using non-convex and non-negative sparse coding is superior than the original ScSPM[1] on several typical databases.  ...  NScSPM is the non-negative sparse model which only use the non-negative constraint, not use the non-convex. 15 Scene Data Set Scene 15 contains 15 categories and 4485 images in all, with 200 to 400 images  ... 
arXiv:1504.06897v1 fatcat:aqmvzqxusvgazharqefdlkekmm

ShopSign: a Diverse Scene Text Dataset of Chinese Shop Signs in Street Views [article]

Chongsheng Zhang and Guowen Peng and Yuefeng Tao and Feifei Fu and Wei Jiang and George Almpanidis and Ke Chen
2019 arXiv   pre-print
captured in different scenes, from downtown to developing regions, using more than 50 different mobile phones. (3) difficulty: the dataset is very sparse and imbalanced.  ...  To illustrate the challenges in ShopSign, we run baseline experiments using state-of-the-art scene text detection methods (including CTPN, TextBoxes++ and EAST), and cross-dataset validation to compare  ...  In the literature, early approaches to scene text detection use "low-level" features to localize texts in natural scene images.  ... 
arXiv:1903.10412v1 fatcat:rbjprketzre7jffltptpijlcne

Arbitrary shape natural scene text detection method based on soft attention mechanism and dilated convolution

Xiao Qin, Jianhui Jiang, Chang-an Yuan, Shaojie Qiao, Wei Fan
2020 IEEE Access  
Based on the aforementioned technique, the proposed method an effectively handle the problem of sparse arranged arbitrary natural scene text detection.  ...  In addition, Jaccard coefficient is used as loss function to promote the post-processing capability of detecting sparse-arranged and arbitrary shape text.  ...  The aforementioned loss functions are often used for large objects detection. However, in natural scene images, texts are usually sparsely arranged and sometimes it only occupies a very small area.  ... 
doi:10.1109/access.2020.3007351 fatcat:vaum76akpzfbxixrqqggppri6m

Multiview Reconstruction of Complex Organic Shapes

Jasenko Zivanov, Thomas Vetter
2015 Procedings of the British Machine Vision Conference 2015  
This dense depth map is then used to detect the occluding contours of that batch while the original sparse depth map is used to estimate their locations and orientations.  ...  Next, we use the contours detected in all the batches to estimate local quadrics in a discretized volume.  ...  This dense depth map is then used to detect the occluding contours of that batch while the original sparse depth map is used to estimate their locations and orientations.  ... 
doi:10.5244/c.29.157 dblp:conf/bmvc/ZivanovV15 fatcat:a7epo4ybfvbcdf6fu77qvcooea

Scene text recognition in multiple frames based on text tracking

Xuejian Rong, Chucai Yi, Xiaodong Yang, Yingli Tian
2014 2014 IEEE International Conference on Multimedia and Expo (ICME)  
Most previous methods of scene text extraction are developed from a single scene image.  ...  Second, a feature representation of STC is employed from dense sampled SIFT descriptors and Fisher Vector. Third, we collect a dataset for text information extraction from natural scene videos.  ...  Scene text recognition in detected image regions is based on the discriminative feature representation of STC.  ... 
doi:10.1109/icme.2014.6890248 dblp:conf/icmcs/RongYYT14 fatcat:xp5a7dbt25fl3p6xgpzurj3sdu
« Previous Showing results 1 — 15 out of 14,414 results