Static and dynamic human activity recognition with VGG-16 pre-trained CNN model
Main Article Content
Human Activity Recognition has been widely studied using the Convolutional Neural Network (CNN) algorithm to classify a person's movements by utilizing data from devices that record movements such as cameras. The benefits generated by this technology are useful for modern devices such as Virtual Reality and Smart Home technology with CCTV cameras. The VGG-16 (Visual Geometric Group with 16 Layers) pre-trained model is one of the models used for transfer learning and has won the Image Net competition. In this study, the authors tested the performance of the VGG-16 model to identify two types of human activity, namely Static and Dynamic. This study uses 1,680 public datasets which are divided into 80% Data Train, 10% Data Validation, and 10% Data Test I. In addition, there are also 100 local datasets as Data Test II. There is no overfitting issue in the training and testing process. The accuracy of the Testing process with public and local images dataset produces a high accuracy of 98.8% and 97% respectively.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work