Text line segmentation using Viterbi algorithm for the palm leaf manuscripts of Dai

Ge, Peng; Yu, Pengfei; Li, Haiyan; He, Lesheng

doi:10.1109/icalip.2016.7846561

Cited by 9 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To segment palm leaf manuscripts of Dai, Ge Peng et al [12] used algorithm based on HMM, to evaluate all segmentation paths. In another study, Quang Nhat and Lee [13] proposed multilingual text line segmentation approach, in which trained fully conventional network (FCN) is used to figure out text lines pattern.…”

Section: Related Workmentioning

confidence: 99%

An Efficient Skewed Line Segmentation Technique for Cursive Script OCR

Malik

Sajid

Ahmad

et al. 2020

Scientific Programming

View full text Add to dashboard Cite

Segmentation of cursive text remains the challenging phase in the recognition of text. In OCR systems, the recognition accuracy of text is directly dependent on the quality of segmentation. In cursive text OCR systems, the segmentation of handwritten Urdu language text is a complex task because of the context sensitivity and diagonality of the text. This paper presents a line segmentation algorithm for Urdu handwritten and printed text and subsequently to ligatures. In the proposed technique, the counting pixel approach is employed for modified header and baseline detection, in which the system first removes the skewness of the text page, and then the page is converted into lines and ligatures. The algorithm is evaluated on manually generated Urdu printed and handwritten dataset. The proposed algorithm is tested separately on handwritten and printed text, showing 96.7% and 98.3% line accuracy, respectively. Furthermore, the proposed line segmentation algorithm correctly extracts the lines when tested on Arabic text.

show abstract

Section: Related Workmentioning

confidence: 99%

An Efficient Skewed Line Segmentation Technique for Cursive Script OCR

Malik

Sajid

Ahmad

et al. 2020

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…The connected-component algorithm [7] has good performances in the non-torched field but cannot nicely split the torched character images. Viterbi-based algorithm [8] and the dripping segmenting algorithm [9] had done well in this task, but both were prone to unsolvable over-segmentation problems. Some other attempts have been made to address this problem by using language model-based recognition algorithms [10][11][12][13].…”

Section: Introductionmentioning

confidence: 99%

Segmentation-free Recognition Algorithm Based on Deep Learning for Handwritten Text Image

Peng

2024

JAIT

View full text Add to dashboard Cite

Segmentation-based offline handwritten character recognition algorithms suffered from the segmenting difficulty of interleaving and touching in handwritten manuscripts. To tackle the problem, a segmentation-free recognition algorithm based on deep learning network is proposed in this paper. The network consists of four neural layers, including input layer for image preprocessing, CNNs layer for feature extraction, BDLSTM layer for sequence prediction, and connectionist temporal classification layer for text sequence alignment and classification. Besides, a novel data processing method is performed for data length equalization. Based on this, groups of experiments, based on six typical databases, involved in evaluation indicators of character correct rate, training time cost, storage space cost and testing time cost are carried out. The experimental results show that the proposed algorithm has better performances in accuracy and efficiency than other classical algorithms.

show abstract

“…The grouping is accomplished by twodimensional voting process that is applied to extract the curve saliency values and normal vectors to construct the character sequences. While most methods employ heuristic approaches to group the local units, the graph-based strategies [6,7], based on stable path or hidden Markov model, are applied to construct nonlinear segmentation paths in document images with dense text lines, like palm leaf manuscripts.…”

Section: Introductionmentioning

confidence: 99%