Scientific and Technical Journal


ISSN Print 2221-3937
ISSN Online 2221-3805

Textareas (TA) detection in images with complex inhomogeneous background is complicated by the fact that the texts in these images are often not separated from other information clearly, and text is a part of this information. Despite the abundance of publications, currently there is no method and system of text areas detection, guaranteed to solve this problem with an acceptable quality. The article describes the developed information technology of text localization in images with complex background. The suggested by the authors' method of text areas localization in images with complex background, based on convolutional neural network, is a basis of proposed information technology. The method takes into account the image's multi-scale decomposition on wavelet basis and training on basis of a characters probabilistic model.

Using models generated in forming the training samples decreases dependence of the localization quality of the number and characteristics of real-world images containing TA, increases flexibility to create the training set, increases localization accuracy after training convolution neural network.

Performance testing of proposed information technology of text localization was conducted. Localization accuracy then on the image of the training sample was 99.9% and in the control sample - 87.7%.

  1. Andrianov A.I. Lokalizatsiya teksta na izobrazheniyah slozhnyih graficheskih stsen,

[Text Localization in Images of Complex Graphical Scenes], (2013), Sovremennyie Problemyi Nauki i Obrazovaniya, 3. URL: Russian).

  1. Coates A., Carpenter B., Case C., Satheesh S., Suresh B., Wang T., Wu D., and Ng A.,(2011), Text Detection and Character Recognition in Scene Images with Unsupervised Feature Learning, The 11th International Conference on Document Analysis and Recognition (ICDAR), pp. 440 – 445.
  2. Epshtein B., Ofek E. and Wexler Y. (2010), Detecting Text in Natural Scenes with Stroke Width Transform, 23rd IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Vol.V.
  3. Kunishige Y., Yaokai F., and Uchida S. (2011), Scenery Character Detection with Environmental Context,The 11th International Conference on Document Analysis and Recognition (ICDAR), pp. 1049 – 1053.
  4. Uchida S., Shigeyoshi Y., Kunishige Y., and Yaokai F.,(2011), A Keypoint-Based Approach Toward Scenery Character Detection,The 11th International Conference on Document Analysis and Recognition (ICDAR), pp. 819 –823.
  5. Du Y., Ai H., and Lao S. (2011), Dot Text Detection Based on FAST Points,The 11th International Conference on Document Analysis and Recognition (ICDAR), pp. 435 – 439.
  6. Jung C., LiuQ.F., and Kim J. Accurate Text Localization in Images Based on SVM Output Scores, 2009. Image and Vision Computing,vol. 27, pp. 1295–1301.
  7. Antoschuk S.G., Nikolenko A.O., Babilunga O.Yu., and Tkachenko O.V. Adaptivna lokalizatsiya simvolnih napisiv na zobrazhennyah metodom veyvlet-analizu, [Adaptive Localization of Character Information at the Images by Wavelet-Analysis Method], (2008), Visnik Zhitomirskogo Tehnol. Un-tu, 4(47), pp. 124 – 130 (InUkrainian).
  8. Nikolenko A.A., and Nguyen Tien T. K. Obnaruzhenie tekstovyih oblastey v video-posledovatelnostyah, [TextRegionsDetectioninVideoFrames], (2012), Iskusstvennyiy Intellekt, 4, pp. 227 – 234 (InRussian).
  9. Delakis М., andGarcia Сr.,  (2008), Text Detection with Convolutional Neural Networks, International Conference on Computer Vision Theory and Applications, pp. 290 – 294.
  10. NikolenkoA.O., Babilunga O.Yu., and Nguyen Tien T.K. Metod vyiavlennia tekstovykh oblastei v videoposlidovnosti z vykorystanniam veivlet-peretvorennia, [Met-hodofDetectingTextareasinVideoSequencesusingWaveletTransform], (2012), XI Vseukrainska Mizhnarodna Konferentsiia “Obroblennia Ssyhnaliv i Zobrazhen ta Rozpiznavannia Obraziv” (UkrOBRAZ’2012), Zbirnyk Naukovykh Prats,Kiev, Ukraine, MNNTsITS, pp. 99 – 102. (InUkrainian)
  11. Nikolenko A.A., Babilunga O.Yu., and Nguyen Tien T.K. Lokalizatsiya tekstovyih oblastey na izobrazheniyah s ispolzovaniem svertochnoy neyronnoy seti, [LocalizationoftheTextareaontheImagesusingaConvolutionNeuralNetwork], (2013), Visnik Natsionalnogo Tehnichnogo Universitetu Harkivskiy Politehnichniy Institut”. Zbirnik Naukovih Prats. Seriya: Informatika ta Modelyuvannya, Harkiv: NTU HPI,19 (992), pp. 121–127 (In Russian).
  12. Nguyen Tien T.K. Obnaruzhenie i raspoznavanie tekstov na izobrazheniyah slozhnyih graficheskih stsen s pomoschyu svertochnoy neyronnoy seti, [TextDetectionandCharacterRecognitioninImagesofComplexGraphicScenesusingConvolutionNeuralNetworks], (2014), Elektrotehn. ta Komp’yut. Sistemi, 13 (89), pp.125 – 130 (InRussian).
  13. Antoschuk S.G., NIkolenko A.A., and Nguyen Tien T.K. Metodlokalizatsiyi teksta na izobrazheniyi s obucheniem na osnove veroyatnostnoy modeli simvolov, [Text Localization Method on Image with Training Based on Characters Probabilistic Models], (2014), Vymiryuvalna ta Obchislyuvalna Tekhnika v Tekhnolo-gichnykh Protsesakh, No.3, pp. 79 – 84 (In Russian).
  14. Vezhnevets V., Otsenka kachestva ra-botyi klassifikatorov, [ClassifierPerformanceEvaluation], (2006), Kompyuternaya Grafika i Multimedia, 2006, 4.

URL: content/view/106(In Russian).

Last download:
8 Feb 2020

[ © KarelWintersky ] [ All articles ] [ All authors ]
[ © Odessa National Polytechnic University, 2014-2018. Any use of information from the site is possible only under the condition that the source link! ]