Cross-Modality Interaction Network for Equine Activity Recognition Using Imbalanced Multi-Modal Data
| Autor: | Endai Huang, Weitao Xu, Axiu Mao, Kai Liu, Haiming Gan, Rebecca S. V. Parkes |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Computer science
Wearable computer wearable sensor TP1-1185 intermodality interaction Biochemistry Convolutional neural network Article Analytical Chemistry Activity recognition class-balanced focal loss Units of measurement Interaction network Animals Horses Electrical and Electronic Engineering Instrumentation Modality (human–computer interaction) Recall business.industry Deep learning Chemical technology deep learning Pattern recognition Atomic and Molecular Physics and Optics equine behavior Neural Networks Computer Artificial intelligence business Algorithms |
| Zdroj: | Sensors, Vol 21, Iss 5818, p 5818 (2021) Sensors (Basel, Switzerland) Sensors Volume 21 Issue 17 |
| ISSN: | 1424-8220 |
| Popis: | With the recent advances in deep learning, wearable sensors have increasingly been used in automated animal activity recognition. However, there are two major challenges in improving recognition performance—multi-modal feature fusion and imbalanced data modeling. In this study, to improve classification performance for equine activities while tackling these two challenges, we developed a cross-modality interaction network (CMI-Net) involving a dual convolution neural network architecture and a cross-modality interaction module (CMIM). The CMIM adaptively recalibrated the temporal- and axis-wise features in each modality by leveraging multi-modal information to achieve deep intermodality interaction. A class-balanced (CB) focal loss was adopted to supervise the training of CMI-Net to alleviate the class imbalance problem. Motion data was acquired from six neck-attached inertial measurement units from six horses. The CMI-Net was trained and verified with leave-one-out cross-validation. The results demonstrated that our CMI-Net outperformed the existing algorithms with high precision (79.74%), recall (79.57%), F1-score (79.02%), and accuracy (93.37%). The adoption of CB focal loss improved the performance of CMI-Net, with increases of 2.76%, 4.16%, and 3.92% in precision, recall, and F1-score, respectively. In conclusion, CMI-Net and CB focal loss effectively enhanced the equine activity classification performance using imbalanced multi-modal sensor data. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|