The A2iA-Telecom ParisTech-UOB System for the ICDAR 2009 Handwriting Recognition Competition

Online Handwritten Chinese/Japanese Character Recognition

“…Kermorvant et al constructed a neural network to combine the top rank candidates of three word recognizers [37]. The two works used the discriminative methods to estimate the combination parameters.…”

“…HMMs were first described in a series of statistical papers [51] and applied to speech recognition [52][53] in the middle of the 1970s. Then, they were applied widely to online handwriting [12][13][14][15][16][17][18][19][20][21] and offline word recognition [32][33][34][35][36][37][38].…”

“…However, it is still an HMM-based method, although it uses the neighborhood relationships of the recognition. Based on the advantages of MRFs, we can assume that applying MRFs instead of HMMs to integrate the information among the neighboring feature vectors can improve performance of offline English or other western word recognition using segmentation-free methods [32][33][34][35][36][37][38].…”

“…Segmentation-free offline word recognition methods using un-structural features in sliding windows combined with hidden Markov models (HMMs) [32][33][34][35][36][37][38] and those using over-(a) Segmentation-free method (b) Over-segmentation-based method segmentation-based methods [39][40][41][42][43] can also be considered as un-structural methods since they use un-structural features.…”

“…A feasible way to overcome the ambiguity of segmentation is called integrated segmentation and recognition, which is classified into segmentation-free and over-segmentation-based methods [31,32] as shown in Figure 1. Segmentation-free methods, mostly combined with hidden Markov model (HMM)-based recognition [20,21,[33][34][35][36][37][38], simply slice the word pattern into frames (primitive segments) and label the sliced frames, which are concatenated into characters during recognition. Such methods do not sufficiently incorporate character shape information.…”