Accelerometer, Gyroscope, and Magnetometer Sensors based Data for Recognizing Handwritten Digits
This dataset consists of sensory data of digits, i.e., from 0 to 9. The dataset is collected from 20 volunteers by using a 9−axis Inertial Measurement Unit (IMU) equipped marker pen. The objective of this dataset is to design classification algorithms for recognizing a handwritten digit in real-time.
The dataset contains Multivariate Time Series (MTS) data of digits, which is collected from a sensors-equipped marker pen. A 9-Axis IMU sensor, attached to the marker pen, is used to record the hand
movements of the writer while writing a digit on a wall-mounted whiteboard. The IMU consists of three sensors: 3-axis accelerometer, 3-axis gyroscope, 3-axis magnetometer. These sensors are connected to a NodeMCU which transfers the data to an MQTT server over Wi-Fi. The sensory data is collected from 20 volunteers at the sampling rate 75 Hz. A writing activity is designed which consists of 5 rows where each row has 10 digits in ascending order [i.e., 0, 1, · · ·, 9]. Therefore, a single writing activity provides an MTS of 50 digits (i.e., each digit 5 times), which is later segmented to have a separate MTS for each of 50 digits. As each volunteer performs the writing activity 10 times, total 50x10x20 (=10000) labeled instances are created in the dataset.
Dataset FilesLOGIN TO ACCESS DATASET FILES
Open Access dataset files are accessible to all logged in users. Don't have a login? Create a free IEEE account. IEEE Membership is not required.