A Comparative Study of Shape and Texture Features for Finger Spelling Recognition in Big Data Applications

With a wide variety of big data applications, Sign Language Recognition has become one of the most important research areas in the field of human-computer interaction. Despite recent progresses, the task of classifying finger spelling is still very challenging in Sign Language Recognition. The visually similarity of some signs, the invisibility of the thumb and the large amount of variation by different signers are all make the hand shape recognition very challenging. The work presented in this paper aims to evaluate the performance of some state-of-the-art features for static finger spelling of alphabets in sign language recognition. The comparison experiments were implemented and tested using two popular data sets. Based on the experimental results, analysis and recommendations are given on the efficiency and capabilities of the compared features.

Thai Finger-Spelling Recognition Using a Cascaded Classifier Based on Histogram of Orientation Gradient Features

Hand posture recognition is an essential module in applications such as human-computer interaction (HCI), games, and sign language systems, in which performance and robustness are the primary requirements. In this paper, we proposed automatic classification to recognize 21 hand postures that represent letters in Thai finger-spelling based on Histogram of Orientation Gradient (HOG) feature (which is applied with more focus on the information within certain region of the image rather than each single pixel) and Adaptive Boost (i.e., AdaBoost) learning technique to select the best weak classifier and to construct a strong classifier that consists of several weak classifiers to be cascaded in detection architecture. We collected 21 static hand posture images from 10 subjects for testing and training in Thai letters finger-spelling. The parameters for the training process have been adjusted in three experiments, false positive rates (FPR), true positive rates (TPR), and number of training stages (N), to achieve the most suitable training model for each hand posture. All cascaded classifiers are loaded into the system simultaneously to classify different hand postures. A correlation coefficient is computed to distinguish the hand postures that are similar. The system achieves approximately 78% accuracy on average on all classifier experiments.