A Method for Determining the Moment of Touching a Card Using Wrist-worn Sensor in Competitive Karuta

Abstract

Abstract:Competitive karuta is an official Japanese card game and is described as "martial art on the tatami." Recently, competitive karuta has attracted a great deal of attention among young people. One of characteristic rules of competitive karuta is that there is no referee; therefore players must judge themselves even if the difficult situation arises. Consequently, the players sometimes get into an argument over their judgement, which disrupts the other matches in the room because all the matches proceed in parallel. In this paper, we propose a system that judges the player who took a card first in a competitive karuta match. Our system measures motion data when players take a card by using a wrist-worn accelerometer and gyroscope, and estimates the times when the players touched the card. From the evaluation experiments, 69.2% of rounds were estimated without error and 99.0% of rounds were estimated within 20-ms error. When our system was introduced on the close game, the accuracy of deciding the player taking a card was 75%.

KARATECH: A Practice Support System Using an Accelerometer to Reduce the Preliminary Actions of Karate

“…The evaluation results showed that it correctly identified the start of a gesture with an accuracy of 87% and classified the type of gesture with an accuracy of 96% for 12 hand gestures. Yamada et al developed a method for estimating the moment of acquiring a card in competitive karuta for contestants wearing an accelerometer and a gyroscope on their wrists [30]. They also improved the method for estimating the time of acquiring cards and applied it to actions other than competitive karuta, detecting the timing of the occurrence of arbitrary actions during various gestures [31].…”

“…This feature makes it difficult to detect only the pre-action part in a series of gestures. Next, as described in Sections 2.3 and 2.4, the pre-action is included during the gesture; therefore, it is difficult to detect the pre-action using methods that detect the point at which the gesture switches in a continuous motion, such as automatic segmentation [23][24][25][26][27][28] or detection of motion occurrence timing [29][30][31]. Finally, the waveform of pre-action acceleration data in the forefist punch is not well-defined, as there are various motions such as fist pulling, arm lowering, shoulder raising, etc., so it is difficult to construct a dataset labeled for each type of pre-action.…”