Human activity recognition (HAR) remains a challenging yet crucial problem to address in computer vision. HAR is primarily intended to be used with other technologies, such as the Internet of Things, to assist in healthcare and eldercare. With the development of deep learning, automatic high-level feature extraction has become a possibility and has been used to optimize HAR performance. Furthermore, deep-learning techniques have been applied in various fields for sensor-based HAR. This study introduces a new methodology using convolution neural networks (CNN) with varying kernel dimensions along with bi-directional long short-term memory (BiLSTM) to capture features at various resolutions. The novelty of this research lies in the effective selection of the optimal video representation and in the effective extraction of spatial and temporal features from sensor data using traditional CNN and BiLSTM. Wireless sensor data mining (WISDM) and UCI datasets are used for this proposed methodology in which data are collected through diverse methods, including accelerometers, sensors, and gyroscopes. The results indicate that the proposed scheme is efficient in improving HAR. It was thus found that unlike other available methods, the proposed method improved accuracy, attaining a higher score in the WISDM dataset compared to the UCI dataset (98.53% vs. 97.05%).

There are a number of issues related to the development of biometric authentication systems, such as privacy breach, consequential security and biometric template storage. Thus, the current paper aims to address these issues through the hybrid approach of watermarking with biometric encryption. A multimodal biometric template protection approach with fusion at score level using fingerprint and face templates is proposed. The proposed approach includes two basic stages, enrollment stage and verification stage. During the enrollment stage, discrete wavelet transform (DWT) is applied on the face images to embed the fingerprint features into different directional sub-bands. Watermark embedding and extraction are done by quantizing the mean values of the wavelet coefficients. Subsequently, the inverse DWT is applied to obtain the watermarked image. Following this, a unique token is assigned for each genuine user and a hyper-chaotic map is used to produce a key stream in order to encrypt a watermarked image using block-cipher. The experimentation results indicate the efficiency of the proposed approach in term of achieving a reasonable error rate of 3.87%.

No abstract