A Japanese Input Method for Mobile Terminals Using Surface EMG Signals

Abstract

Abstract. The common use of mobile terminals is for text input. However, mobile terminals cannot be equipped with sufficient amount of keys because of the physical restrictions. To solve this problem we developed an input method using surface electromyogram (sEMG), treating arm muscle movements as input signals. This method involves no physical keys and can be used to input Japanese texts. In our experiments, the system was capable of inputting Japanese characters with a finger motion recognition rate of approximately 80%.

Early gesture recognition with adaptive window selection employing canonical correlation analysis for gaming

“…It achieved good recognition rates for ten‐digit recognition, but it requires large training data and many pre‐processing filtering. In [5], the authors developed a Japanese text input method for mobile phones using surface electromyogram (sEMG). EMG signals, which detect the activities of related muscles during a gesture performance is more suitable for capturing fine motions such as wrist and finger movements.…”

“…Introduction: Any interactive system depends on extracting discriminating features to train a recognition algorithm so that it can recognise a new movement during testing phase as soon as it starts before its end. According to the sensing technologies used to capture gestures, recent researches can be categorised into two main groups: computer vision-based techniques [1,2] and sensor-based techniques [3][4][5]. Common problems facing vision-based techniques are light and background variations.…”

Real time vision/sensor based features processing for efficient HCI employing canonical correlation analysis

“…Wearable devices of all kinds are becoming popular [11,12]. In [11], the authors presented an accelerometerbased digital pen that uses trajectory recognition algorithm to convert time-series acceleration signals into features vectors.…”

“…On the other hand, elec-tromyography (EMG) signals, which detects the activities of related muscles during a gesture performance are more suitable for capturing fine motions such as wrist and finger movements. In [12] the authors developed a Japanese text input method for mobile phones using surface electromyogram (sEMG). The proposed algorithm employed some signal processing techniques, generic dictionary and learning dictionary.…”