A metasynthetic approach for segmenting handwritten Chinese character strings

Authors:
Zhizhen Liang;Pengfei Shi
Affiliations:
Institute of Image Processing and Pattern Recognition, Shanghai Jiaotong University, Hushan Road 1954, Shanghai 200030, PR China;Institute of Image Processing and Pattern Recognition, Shanghai Jiaotong University, Hushan Road 1954, Shanghai 200030, PR China
Venue:
Pattern Recognition Letters
Year:
2005

Citing 8
Cited 1

Recognition of isolated and simply connected hand-written numerals

Pattern Recognition
Segmenting handwritten Chinese characters based on heuristic merging of stroke bounding boxes and dynamic programming

Pattern Recognition Letters
Recognition-based handwritten Chinese character segmentation using a probabilistic Viterbi algorithm

Pattern Recognition Letters
Segmentation of Single- or Multiple-Touching Handwritten Numeral String Using Background and Foreground Analysis

IEEE Transactions on Pattern Analysis and Machine Intelligence
Offline General Handwritten Word Recognition Using an Approximate BEAM Matching Algorithm

IEEE Transactions on Pattern Analysis and Machine Intelligence
Goal-Directed Evaluation of Binarization Methods

IEEE Transactions on Pattern Analysis and Machine Intelligence
Lexicon-Driven Segmentation and Recognition of Handwritten Character Strings for Japanese Address Reading

IEEE Transactions on Pattern Analysis and Machine Intelligence
Integrated segmentation and recognition of handwritten numeralswith cascade neural network

IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews

Segment confidence-based binary segmentation (SCBS) for cursive handwritten words

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.10

Visualization

Abstract

In this paper, a metasynthetic method is proposed to segment handwritten Chinese character strings. The Viterbi algorithm is firstly applied to search segmentation paths and several rules are used to remove redundant paths. Then a background-thinning method is further adopted to obtain non-linear segmentation paths. If there are not touching characters, a dynamic programming algorithm is applied to merge components. For touching characters, we apply background and foreground information to obtain candidate segmentation paths and the feature vectors are constructed in terms of peripheral features. Then the mixture probabilistic density function whose parameters are obtained by the EM algorithm is used to choose the best segmentation path. Experimental results demonstrate that the proposed scheme effectively segments handwritten Chinese characters and achieves an improvement over previous methods.