Emotion recognition has important applications in human-computer interaction. Various sources such as facial expressions and speech have been considered for interpreting human emotions. The aim of this paper is to develop an emotion recognition system from facial expressions and speech using a hybrid of machine-learning algorithms in order to enhance the overall performance of human computer communication. For facial emotion recognition, a deep convolutional neural network is used for feature extraction and classification, whereas for speech emotion recognition, the zero-crossing rate, mean, standard deviation and mel frequency cepstral coefficient features are extracted. The extracted features are then fed to a random forest classifier. In addition, a bi-modal system for recognising emotions from facial expressions and speech signals is presented. This is important since one modality may not provide sufficient information or may not be available for any reason beyond operator control. To perform this, decision-level fusion is performed using a novel way for weighting according to the proportions of facial and speech impressions. The results show an average accuracy of 93.22 %.
The automatic liquid filling system is used in different applications such as production of detergents, liquid soaps, fruit juices, milk products, bottled water, etc. The automatic bottle filling system is highly expensive. Where, the common filling systems required to complex changes in hardware and software in order to modify volume of liquid. There are many important variables in the filling process such as volume of liquid, the filling time, etc. This paper presents a new approach to develop an automatic liquid filling system. The new proposed system consists of a conveyor subsystem, filling stations, and camera to detect the level of the liquid at any instant during the filling process. The camera can detect accurately the leve
... Show MoreThis paper proposes and tests a computerized approach for constructing a 3D model of blood vessels from angiogram images. The approach is divided into two steps, image features extraction and solid model formation. In the first step, image morphological operations and post-processing techniques are used for extracting geometrical entities from the angiogram image. These entities are the middle curve and outer edges of the blood vessel, which are then passed to a computer-aided graphical system for the second phase of processing. The system has embedded programming capabilities and pre-programmed libraries for automating a sequence of events that are exploited to create a solid model of the blood vessel. The gradient of the middle c
... Show MoreSensitive information of any multimedia must be encrypted before transmission. The dual chaotic algorithm is a good option to encrypt sensitive information by using different parameters and different initial conditions for two chaotic maps. A dual chaotic framework creates a complex chaotic trajectory to prevent the illegal use of information from eavesdroppers. Limited precisions of a single chaotic map cause a degradation in the dynamical behavior of the communication system. To overcome this degradation issue in, a novel form of dual chaos map algorithm is analyzed. To maintain the stability of the dynamical system, the Lyapunov Exponent (LE) is determined for the single and dual maps. In this paper, the LE of the single and dual maps
... Show MoreTo damp the low-frequency oscillations which occurred due to the disturbances in the electrical power system, the generators are equipped with Power System Stabilizer (PSS) that provide supplementary feedback stabilizing signals. The low-frequency oscillations in power system are classified as local mode oscillations, intra-area mode oscillation, and interarea mode oscillations. A suitable PSS model was selected considering the low frequencies oscillation in the inter-area mode based on conventional PSS and Fuzzy Logic Controller. Two types of (FIS) Mamdani and suggeno were considered in this paper. The software of the methods was executed using MATLAB R2015a package.
This work investigates a simulation model of an underwater optical wireless communication (UOWC) system. Several water scenarios are considered: Harbor I (HA-I), Harbor II (HA-II), Coastal Ocean (CO), Clear Ocean (CL), and Pure Sea (PU). A laser diode (LD) with modulation schemes (NRZ-OOK) transmits data at various speeds of 2.5 Gbps, 5 Gbps, and 10 Gbps. To identify the optical signal, a single-photon detection (SPD), APD and PIN photodiodes are utilized. The analytical evaluation of the performance is executed using Q-factor, received power and bit error rate (BER). According to the results, the PU achieved an underwater distance of 35.5 m, 35 m, 34.5 m, for data tran
Enhancement of the performance for hybrid solar air conditioning system was presented in this paper. The refrigerant temperature leaving the condenser was controlled using three-way valve, this valve was installed after the compressor to regulate refrigerant flow rate towards the solar system. A control system using data logger, sensors and computer was proposed to set the opening valve ratio. The function of control program using LabVIEW software is to obtain a minimum refrigerant temperature from the condenser outlet to enhance the overall COP of the unit by increasing the degree of subcooled refrigerant. A variable load electrical heater with coiled pipe was used instead of the solar collector and the storage tank to simulate the sola
... Show MoreFace recognition and identity verification are now critical components of current security and verification technology. The main objective of this review is to identify the most important deep learning techniques that have contributed to the improvement in the accuracy and reliability of facial recognition systems, as well as highlighting existing problems and potential future research areas. An extensive literature review was conducted with the assistance of leading scientific databases such as IEEE Xplore, ScienceDirect, and SpringerLink and covered studies from the period 2015 to 2024. The studies of interest were related to the application of deep neural networks, i.e., CNN, Siamese, and Transformer-based models, in face recogni
... Show More