- Face and Expression Recognition
- Speech and Audio Processing
- Robotic Path Planning Algorithms
- Speech Recognition and Synthesis
- Indoor and Outdoor Localization Technologies
- Face recognition and analysis
- IoT-based Smart Home Systems
- Music and Audio Processing
- Robotics and Sensor-Based Localization
- Image and Video Stabilization
- Advanced Algorithms and Applications
- Underwater Vehicles and Communication Systems
- Image Retrieval and Classification Techniques
- Video Surveillance and Tracking Methods
- E-commerce and Technology Innovations
- Robotics and Automated Systems
- Smart Parking Systems Research
- Autonomous Vehicle Technology and Safety
- Muscle activation and electromyography studies
- Cloud Computing and Resource Management
- Currency Recognition and Detection
- Antenna Design and Optimization
- Power Quality and Harmonics
- Time Series Analysis and Forecasting
- Embedded Systems and FPGA Design
University of Technology - Iraq
2015-2024
University of Technology
2021
Middle Technical University
2020
This paper describes the different classifier methods with minimum means of clusters to achieve face recognition rate humans from feature extracted training image data for many sets images as a base. Principal Component Analysis (PCA) is robust method used extraction techniques but decreases variation person's actions. The features are light insensitive, individual, hidden, and activity effective biometric recognition. treats two dimensions problems, fact take advantage these human faces...
A number of studies have focused on shortest-path algorithms as a way to pick the most feasible route for movement task in order reduce both expense and time complexity. To resolve various difficulties, designers commonly apply with least or expense; nonetheless, certain difficulties present greater challenges. Dijkstra's technique was thus applied specific example utilising two alternative digraphs representing one-way two-way street routes. Every set be adjusted conditions changed, and, by...
This paper introduces a smart mobile robot for tracking an optimal path road network. The development of robots became more useful and efficient in wide range areas. Therefore, the proposed system was designed by using microcontroller type Raspberry Pi 3 with associate Python language implemented on known networks based Dijkstra's algorithm to choosing best reach target robot. has putted solution shortest length between two nodes graphic search. experimental results were taken from gradient...
Abstract The widely applied location-based services require a high standard for positioning technology. With the development of wireless communication technology, based on Wi-Fi are increasingly demanded. To understand accuracy limits under certain given Access Point topology and to be able make design recommendations. This paper proposed method proper deployment AP devices building electrical department in university technology computing measurements Received Signal Strength (RSS)....
This paper presents an efficient face recognition system based on Hidden Markov Model (HMM) and the simplest type “Haar” of Discrete Wavelet Transform (DWT). The one dimensional ergodic HMM with Gaussian outputs, which represent robust HMM, is used in proposed work. A novel method introduced for selecting training images implemented by choosing that have odd identifying numbers from database. Some these are replaced according to trial-and-error results. work achieves maximum rate (100%),...
this paper introduces a modified management system for smart electrical energy meter. The proposed meter has feature to inform the consumer about consumption, serves consumption reduction by self-awareness of real consumption. other important is eliminating difficulties direct contact employee and getting reading conventional also error in bill. proposal consists two parts: client's side part (the prepaid meters houses) server (electrical sub-station). client an RFID reader, credit cards,...
This paper introduced an approach to design and implement embedded SoPC (System on a Programmable Chip) technique with Altera Nios II processor for real-time speech recognition system by developing hardware/software minimum usage of resources (hardware components) relatively small size software reduce memory utilization. is achieved using Mel Frequency Cepstral Coefficients (MFCCs) as signal feature extraction (observation vector). Using Gaussian Mixture Model (GMM) model the observation...
The huge ubiquitous deployment of wireless technologies and the pervasive availability smart devices, indoor localization has become one most interesting topics in numerous location-based services. Where several methods have been adopted to estimate localize mobile user location. earliest method was based on received signal strength (RSS) measurement. However, such showed weakness handling accurate estimation results. Recently, researchers start gather different achieve advantages each...
In personal image recognition algorithms, two effective factors govern the system’s evaluation, rate and size of database. Unfortunately, proportional to increase in training sets. Consequently, that increases processing time memory limitation problems. This paper’s main goal was present a robust algorithm with minimum data sets high rate. Images for ten persons were chosen as database, nine images each individual full version set, one person out set test pattern before database reduction...
It began to rely a lot on face detection and recognition techniques use them in various applications our daily lives, especially recent years, which motivated companies researchers develop this technology reduce its errors. Despite that, the results of these are still affected by many factors, most notably age, accuracy, difference illumination or low lightness images. For reason, we will create system that reduces effect different image increases using Eigenface algorithm (or PCA) with...
This paper presents the performance investigation of three phase shunt active power filter (SAPF) using proportional integral (PI) and Fuzzy Logic controller (FLC). The control designing applied to filters (APFs) play a key role on improvement APF. Compensate harmonic component for supply current, can improve quality enhance reliability stability utility. In classical systems like PI controller, information controlled system is essential in formulation set algebraic differential equations,...
With the rapid development in field of communication networks, an indoor positioning system (IPs) has become extremely popular recent years. In this paper, we compute access points (AP’s) localization to get best coverage building that takes as a case study and administration o
This paper presents a study on finding an optimal wavelet filter for denoTsing surface electromyography signal, the signal was captured from biceps brachii muscle of human arm, stored in one-dimensional matrix and conducted series procedures to reduce noise. The performance has been tested based upon nearest five filters terms shape form original after subjected three noisy Gaussian environments at different noise ratio. A tremendous amount results obtained, These show that fourth order...
This paper introduced an approach to design and implement a system control opening of security door by means the human voice recognize password. The aim this is speaker word open obtain more high recognition rate. proposes Mel-Frequency Cepstral Coefficient (MFCC) technique as feature extraction combined with its first second derivative including power computation speech frames that yields total multi-dimensional features, classification or features matching using Dynamic Time Warping (DTW)...
This paper presents an approach to study the effect of different eigenfaces levels on faces recognition rate using principal component analysis. The increase in strength variables and lighting facial geometry represent human face , has been analysis (PCA) image whole . is a statistical measurement method which works field linear can be used reduce dimensions thus serve calculations significantly database It gives better accuracy higher experiment was conducted 50 images from (ORL), 40 for...
Speech recognition is widely used in robot control and automation. Nevertheless, the use of speech robots limited due to its susceptibility background noise. This paper proposes a algorithm noisy environments. The proposed based on Perceptual Linear Predictive Cepstral Coefficients (PNCC), which noise-resistant feature extraction technique, Modified K-Nearest Neighbors (KNN) with Dynamic Time Warping (DTW) as classifier. A new KNN-DTW classifier proposed, integrating weighted KNN DTW....
Robot navigation involves a challenging task: path planning for mobile robot operating in changing environment. This work presents an enhanced Q-learning based technique. For robots dynamic environments, algorithm and few heuristic searching techniques are suggested. Enhanced employs novel exploration approach that blends Boltzmann ε-greedy exploration. Heuristic also offered order to constrict the orientation angle variation range narrow search space. In meantime, robotics literature of...
In engineering, the use of mobile robots to teach automatic control is becoming more common because interesting experiments that can be conducted with them. this paper, a robot applies reinforcement learning in different scenarios shown, get rewards, agent learns by acting environment. creating balance between new information and our current understanding way, algorithm divided into two stages: stage operational stage. first phase, how go from where it known destination, builds matrix...
This paper presents the effect of discrete wavelet decomposition on database reduction, which is used in biometric face recognition. The Olivetti and Oracle Research Laboratory own databases for training testing. first level was applied to testing faces made size decreasing by 75% out original size. Both are treated principal component analysis algorithm get that recognition rate. Principal experiment results appeared reduction rate between reduced 2%. approach more efficient database,...
This paper introduce an approach to study the effects of different levels environment noise on recognition rate speech systems, which are not used any type filters deal with this issue. is achieved by implementing embedded SoPC (System a Programmable Chip) technique Altera Nios II processor for real-time system. Mel Frequency Cepstral Coefficients (MFCCs) was signal feature extraction (observation vector). Model observation vector voice information using Gaussian Mixture (GMM), model passed...