Arbitrary oriented multilingual text detection and segmentation using level set and Gaussian mixture model
- PDF / 2,101,087 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 74 Downloads / 248 Views
SPECIAL ISSUE
Arbitrary oriented multilingual text detection and segmentation using level set and Gaussian mixture model H. T. Basavaraju1 · V. N. Manjunath Aradhya1 · M. S. Pavithra2 · D. S. Guru3 · Vikrant Bhateja4,5 Received: 9 May 2020 / Revised: 29 July 2020 / Accepted: 8 August 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract A pioneer conceptual combination of level set method and Gaussian Mixture Model (GMM) is presented in the described multilingual, arbitrary-oriented character segmentation. The method is a serial accomplishment processes of Gaussian low pass filter, single level 2 Dimensional Discrete Wavelet Transform (2D DWT) for better feature extraction and implemented level set method, k-means clustering algorithm with GMM to achieve a veracious character segmentation results by detecting great measure of true text region of an image. The proposed method segments a character chiefly by distinguishing the touching character constituents and also deals discontinuities presence in a character by using Laplacian of Gaussian filter and morphological bridge function in a intellectual way. The exhibited method was explored on Multi-script Robust Reading Competition dataset and on our privately collected graphical and handwritten multilingual, arbitrarily-oriented text images. The suggested method is compared with the well known multilingual and arbitrarily-oriented charter segmentation methods. The described method attains better segmentation outcomes when compared to the familiar functioning methods. Hence, the suggested method is highly suitable to consider as an improved, standard and procedural technique. Keywords Gaussian low pass filter · Single level 2D DWT · Level set method · k-Means algorithm · LOG filter · Gaussian mixture model(GMM) * H. T. Basavaraju [email protected]
1 Introduction
V. N. Manjunath Aradhya [email protected]
Text present in the real world environment is captured by digital camera, hence the large number of text based images and videos are deposited in the multimedia library. This multimedia library is facing difficult to store and retrieve the respective data in the database. Therefore, an effective and automatic indexing and retrieval algorithm needs to be developed. Text present in image or video frame helps to build a proficient algorithm for effective indexing and retrieval process. Character recognition is the process of recognizing text characters in the form of electronic files. With character recognition, text-search has achieved in the digital domain, with this content-based image indexing and retrieval has been executing across the digital domain [1]. In a process of identifying the characters, Optical Character Recognition (OCR) [2] is performed with the optical properties such as refraction and the text object’s refraction index. Many of the OCR engines have been developed to different domain-specific OCR applications like as receipt OCR, invoice OCR, check OCR, legal billing document OCR. These can be used for data e
Data Loading...
