The Effect of Axis-Wise Triaxial Acceleration Data Fusion in CNN-Based Human Activity Recognition

Abstract

The triaxial accelerometer is one of the most important sensors for human activity recognition (HAR). It has been observed that the relations between the axes of a triaxial accelerometer plays a significant role in improving the accuracy of activity recognition. However, the existing research rarely focuses on these relations, but rather on the fusion of multiple sensors. In this paper, we propose a data fusion-based convolutional neural network (CNN) approach to effectively use the relations between the axes. We design a single-channel data fusion method and multichannel data fusion method in consideration of the diversified formats of sensor data. After obtaining the fused data, a CNN is used to extract the features and perform classification. The experiments show that the proposed approach has an advantage over the CNN in accuracy. Moreover, the single-channel model achieves an accuracy of 98.83% with the WISDM dataset, which is higher than that of state-of-the-art methods.

IoT System Based on parameter optimization of Deep Learning using Genetic Algorithm

“…For example some researchers used deep learning to improve the performance of HAR systems [12][13][14]. While others focused on enhancing quality of DL by increasing the information in the training phase [15,16] or by adapting the sensor data [17,18]. Also M. Zeng et al [19] preferred to use DL technique after comparing it with traditional machine learning and finding that DL achieved higher accuracy.…”

“…However, there are trends in many studies to suggest that there are two problems in activities confusions [12,[16][17][18][19][56][57]: Group1 (walking, up and downstairs) and Group2 (sit, stand). For more explanation, experiments results of J. Lee et al [57] showed that 6% of ascending activity was incorrectly classified as walking but 5% of walking activity was incorrectly classified as ascending stairs.…”

“…It is not surprising that G-CNN and CNN are once again struggling to recognize a sitting from the standing activity. Many researchers [12,16,17,18] contend that this misclassification could be a common issue.…”