Segmentation techniques for handwritten script recognition system

Mathew, C Jibu; Shinde, Ravi C.; Patil, C. Y.

doi:10.1109/iccpct.2015.7159397

Cited by 3 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Image segmentation aims at minimizing intra-variance and maximizing inter-variance among regions [1]. Several image processing and computer vision applications rely on image segmentation in different fields, such as medical imaging [2], autonomous vehicle navigation [3], and face or optical character recognition [4]. Available image segmentation algorithms in the literature require different levels of supervision to suit different types of applications.…”

Section: Introductionmentioning

confidence: 99%

ALFO: Adaptive Light Field Over-Segmentation

Hamad¹,

Conti

Nunes

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Automatic image over-segmentation into superpixels has attracted increasing attention from researchers to apply it as a pre-processing step for several computer vision applications. In 4D Light Field (LF) imaging, image over-segmentation aims at achieving not only superpixel compactness and accuracy but also cross-view consistency. Due to the high dimensionality of 4D LF images, depth information can be estimated and exploited during the over-segmentation along with spatial and visual appearance features. However, balancing between several hybrid features to generate robust superpixels for different 4D LF images is challenging and not adequately solved in existing solutions. In this paper, an automatic, adaptive, and view-consistent LF over-segmentation method based on normalized LF cues and K-means clustering is proposed. Initially, disparity maps for all LF views are estimated entirely to improve superpixel accuracy and consistency. Afterwards, by using K-means clustering, a 4D LF image is iteratively divided into regular superpixels that adhere to object boundaries and ensure cross-view consistency. Our proposed method can automatically adjust the clustering weights of the various features that characterize each superpixel based on the image content. Quantitative and qualitative results on several 4D LF datasets demonstrate outperforming performance of the proposed method in terms of superpixel accuracy, shape regularity and view consistency when using adaptive clustering weights, compared to the state-of-the-art 4D LF over-segmentation methods.INDEX TERMS Automatic segmentation, adaptive light field over-segmentation, superpixels.

show abstract

Section: Introductionmentioning

confidence: 99%

ALFO: Adaptive Light Field Over-Segmentation

Hamad¹,

Conti

Nunes

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…More than one reason recognizes Arabic language different from other languages according to its shape and the writing style [2]. Here are some of these reasons:…”

Section: Introductionmentioning

confidence: 99%

“…There are some different dialects identified with the Arabic language as these dialects have a few similitudes with it whence the character shapes or from the articulation. These dialects are Jawi, Persian, Bengali, and different dialects [2].…”

Section: Introductionmentioning

confidence: 99%

Arabic Handwriting Recognition Model based on Neural Network Approach

Abdullah¹,

Agal²,

Alharthi³

et al. 2019

IJATCSE

View full text Add to dashboard Cite

Arabic language as the main language of more than millions of people all over the world has attracted researches in the handwriting script recognition field. Arabic scripts have many difficulties which make Arabic Handwriting recognition challenging. In this paper the model proposed aimed at recognizing Arabic words the IFN / ENIT dataset. The model is an OCR using Neural Network classifier preceded by a set of preprocessing techniques includes removing spaces between words, bolding the words, baseline estimation and correction and resizing the words images. The recognition rate of the proposed model is 70%. This result showed how much the proper selection of the preprocessing steps affects the recognition rate of handwritten words. The propo sed model differs from other suggested models in Arabic handwriting field, according, the unlike preprocessing steps applied on a model and a new approach in estimating and correcting words baseline.

show abstract