A5051290262- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Video Surveillance and Tracking Methods
- Image Processing Techniques and Applications
- Infrared Target Detection Methodologies
- Image and Signal Denoising Methods
- Advanced Measurement and Detection Methods
- Neural Networks and Applications
- Advanced Image Processing Techniques
- Medical Image Segmentation Techniques
- Face and Expression Recognition
- Remote-Sensing Image Classification
- Face recognition and analysis
- Image Enhancement Techniques
- Robotics and Sensor-Based Localization
- Image and Video Quality Assessment
- Advanced Data Compression Techniques
- Remote Sensing and Land Use
- Blind Source Separation Techniques
- Power Quality and Harmonics
- Image Retrieval and Classification Techniques
- Image and Object Detection Techniques
- Machine Learning and ELM
- Industrial Vision Systems and Defect Detection
- Color Science and Applications
University of Isfahan
2015-2024
University of Zanjan
2009
Amirkabir University of Technology
2002-2003
Dr. A.P.J. Abdul Kalam Technical University
2001
The Visual Object Tracking challenge VOT2017 is the fifth annual tracker benchmarking activity organized by VOT initiative. Results of 51 trackers are presented; many state-of-the-art published at major computer vision conferences or journals in recent years. evaluation included standard and other popular methodologies a new "real-time" experiment simulating situation where processes images as if provided continuously running sensor. Performance tested typically far exceeds baselines. source...
In this paper, a computer vision-based algorithm for golden delicious apple grading is proposed which works in six steps. Non-apple pixels as background are firstly removed from input images. Then, stem end detected by combination of morphological methods and Mahalanobis distant classifier. Calyx region also applying K-means clustering on the Cb component YCbCr color space. After that, defects segmentation achieved using Multi-Layer Perceptron (MLP) neural network. next step, calyx regions...
In this paper, a new predictive direct torque control (DTC) method is proposed, improving the dynamic response of classical DTC and reducing torque- flux-ripples through voltage vector with an optimal phase. transient state, phase selected in manner where fastest achieved, while steady stator flux amplitude reaches its commanding value at end cycle. applied to motor time duration calculated achieve minimum ripple. The five-segment space-vector modulation used synthesize vector, fixed...
In this paper, a new predictive direct torque control (DTC) method is proposed, which improves the dynamic response of conventional DTC in transient state and yields minimum flux ripples steady through an optimal voltage vector. At state, magnitude, phase, time duration vector are adjusted manner where obtained; whereas parameters fastest achieved. The space-vector modulation used synthesizing selected fixed switching frequency order to improve system efficiency, principle maximum per ampere...
In this paper, a new advanced deadbeat direct torque and flux control (A-DB-DTFC) system is proposed that improves the steady-state transient-state performances of permanent-magnet synchronous motor by adopting two improved methods. Whenever error between its reference value low, an method adopted A-DB-DTFC system, in which phase time duration voltage vector applied to are adjusted manner stator reach their values after just one cycle. high, another where such fastest response achieved....
Building extraction from high-resolution satellite images (HRSI) in urban areas is an intricate problem. Recent studies proposed different methods during 2005–2015. However, HRSI, they have not investigated the effects of challenges altogether. This paper non-building features which are main drawbacks building extraction. To overcome each challenge, it reviews recent strategies between 2005 and 2015. The pros cons strategy discussed, proper combined to generate hybrid methods. Lower cost...
Despite significant advances and innovations in deep network-based vehicle detection methods, finding a balance between detector accuracy speed remains challenge.This study aims to present an algorithm that can manage the of real-time while increasing with comparable high-speed detectors.To this end, Fast-Yolo-Rec is proposed.The proposed method includes new Yolo-based network LSTM-based position prediction networks.The semantic attention mechanism spatial module (SSAM) significantly impacts...
In order to distinguish between benign and malignant types of pigmented skin lesions, computerized procedures have been developed for images taken by different equipment that the most available one them is conventional digital cameras. this research, a new procedure detect melanoma from lesions using macroscopic presented. The are cameras with spatial resolution higher than megapixel considering no constraints special conditions during imaging. proposed procedure, methods weaken effect...
histogram which can be modeled as a mixture of two Gaussian density functions, estimating these densities inpractice is not simply feasible. The objective this paper to use adaptive particle swarm optimization (APSO) forthe suboptimal estimation the means and variances functions; then, computationof optimal threshold value straightforward. comparisons experimental results in wide range complexbimodal images show that proposed thresholding algorithm presents higher correct detection rate...
Getting the advantages of hyperspectral remote sensing images depends on overcoming challenges posed by their large number bands. One approach is selecting appropriate bands metaheuristic methods. Initial versions these methods usually suffer from trapping into local optima, so they do not work optimally in best band subset. The training samples also important selection. With insufficient samples, even improved yield desired result. However, providing sufficient ground data as costly. In...
Automated building extraction from single, very-high-resolution (VHR) images is still one of the most challenging tasks for urban planning, estimating population, understanding dynamics, and many other applications. The complexities objects have caused buildings to be oversegmented into multiple segments in object-based image analysis (OBIA) method. Selecting appropriate segmentation scale parameter a major challenge OBIA that influences discriminative features extraction, especially...
With strong evanescent waves, optical microfibers (MFs) provide guided lights the ability to directly interact with surrounding environments, whereby several fiber optics chemical sensors have been realized. In this study, based on MFs external refractive index (RI) sensitivity, a multimode (MMF) specklegram RI sensor MMF–MF–MMF configuration is presented. As MF Section exposed liquids of different RIs (in range 1.333–1.368), interaction between and waves modulates guidance status MF,...
This paper introduced an experimental evolution of the effectiveness utilizing various moments as pattern features in human face technology. In this paper, we apply pseudo Zernike (PZM) for recognition faces two-dimensional images, and compare their performance with other type moments. The that have used are (ZM), Legendre (LM). We shape information localization, also a radial basis function (RBF) neural network classifier application. classification is dependent on moment order well...