21 August 2023 in IEEE Access
In the elderly population, falls are one of the leading causes of fatal and non-fatal injuries. Fall detection and early alarms play an important role in mitigating the negative effects of falls, especially given the growing proportion of the elderly population. Due to their non-intrusive nature, data availability, and low deployment costs, RGB videos have been used in many previous studies to detect falls. The RGB data, however, can be affected by background environment changes, resulting in non-recognition. To overcome these challenges, many researchers propose extracting skeleton data from RGB videos and using it for fall detection. Although there have been multiple surveys on fall detection, most of them focus on assessing fall detection systems using different kinds of sensors, and a comprehensive evaluation of skeleton-based fall detection in RGB videos is lacking. In this paper, we examine the most recent advances in skeleton-based fall detection in RGB videos, from handcrafted feature-based methods to advanced deep learning algorithms. Further, we present several skeleton-based fall detection techniques and their performance results on various benchmark datasets, along with challenges and future directions in this field.
23 December 2024 in IEEE Access
In today’s digital age, images are vulnerable to manipulation for malicious purposes such as spreading fake news, prompting active research in image forgery detection. With the advances in deep learning (DL), convolutional neural network (CNN) and Transformer models have emerged as prominent tools in this field. However, individual models may excel for certain images while performing poorly for others, leading to variability in model performance. To address this issue, this paper proposes an ensemble approach that combines predictions from multiple models to improve system performance and robustness. First, we utilize a set of pretrained DL models, including CNN-based models, Transformer-based models, and fusion models that combine each of these architectures, and select the best-performing models. Then, these selected models are employed in an ensemble approach using hard voting and soft voting to evaluate their collective performance. Notably, among the selected ensemble of ConvNeXt, SwinV2, and CAFormer, the hard voting technique achieves an accuracy of 97.42%, which is approximately a 6.34% improvement over the baseline model. This result confirms the effectiveness of the proposed ensemble approach for the forgery detection.