hamid shayegh - Academia.edu (original) (raw)

Papers by hamid shayegh

Abstract.This paper is a quick overview of Kaveh 2012 rescue virtual robot team which de-scribes ... more Abstract.This paper is a quick overview of Kaveh 2012 rescue virtual robot team which de-scribes our improvements and works in relation with robotic science and AI in order to partici-pate in Mexico 2012 Robocup Competitions. This year we are going to use some new concepts to dissolve some problems such as Multi Robot Exploration, Localization and mapping (SLAM), Victim Detection and System Architecture.

The change capability of information flow in organizations is a major request that is required fr... more The change capability of information flow in organizations is a major request that is required from IT masters by managers. In this paper we develop a novel method for achieving this goal. This method uses UML activity and sequence diagrams in front-end section, and tree structure of XML language in back-end section. Extended UML diagrams provide an easy designing mode for managers that they can easily design flows and do their favorite changes. In the back-end or business process, graphical diagrams convert to XML trees and thus the process of them can do flexible because the nature of XML language makes a hierarchical tree using nested tags. Using this method can solve the old problem of managers that want a fast, flexible system for supporting policy changes in organizations.

License Plate Recognition (LPR) plays an important role on the traffic monitoring and parking man... more License Plate Recognition (LPR) plays an important role on the traffic monitoring and parking management. A robust and efficient method for enhancing accuracy of license plate characters recognition based on K Nearest Neighbours (K-NN) classifier is presented in this paper. The system first prepares a contour form of the extracted character, then the angle and distance feature information about the character is extracted and finally K-NN classifier is used to character recognition. Angle and distance features of a character have been computed based on distribution of points on the bitmap image of character. In K-NN method, the Euclidean distance between testing point and reference points is calculated in order to find the k-nearest neighbours. We evaluated our method on the available dataset that contain 1200 sample. Using 70% samples for training, we tested our method on whole samples and obtained 99% correct recognition rate.Further, we achieved average 99.41% accuracy using three...

2014 6th Conference on Information and Knowledge Technology (IKT), 2014

This paper represents a new and intelligent system for license plate recognition used in many cas... more This paper represents a new and intelligent system for license plate recognition used in many cases, for example: unattended parking lots, security control of restricted areas, traffic law enforcement, congestion pricing, and automatic toll collection. The proposed system is a process based on a combination of the complement methods that increases the accuracy of output data and is generally accountable on the images that have many problems and aren't answerable with a particular approach. The performance of this system is in a form that, firstly, the input image is given to all the complement methods in order for the location of the plate candidate to be determined, then their response is investigated according to the valence function and finally the location of plate is properly determined by the majority of votes and ensures the accuracy of output data. If the complement methods are chosen correctly, this proposed system will be responsive to the images in which the license plate recognition is likely. This method has been tested on two data sets that have different images of the background, considering the distance, and angle of view so that the correct extraction rate of plate location reached at 100% and 99% respectively.

ICCKE 2013, 2013

ABSTRACT Human detection has recently received significant attention in the field of computer vis... more ABSTRACT Human detection has recently received significant attention in the field of computer vision. Accurate detection of human bodies is an essential component required by a variety of applications such as automated surveillance, advanced user interface and sport analysis. In this paper, we present a new method for human detection in video frames, using learning algorithm, feature extraction and sliding window. People detection is a challenging topic, especially in complex scenes when occlusion occurs. In this paper a new technique based on the body part detection and HOG (Histogram of Gradient) features is proposed for human detection with occlusions. Proposed HOG descriptor provides lower feature vector length in comparison with state of the art methods approaches. Using human detectors alone, cause to high frequent false positive. Thus, we propose a novel classifier-fusion learning algorithm, instead of single classifier. The experimental results show that the new proposed method provides lower false positive and raise higher precision and recall in comparison with state of the art methods.

… Conference on IT to …, 2009

This paper has been withdrawn by the author due to a crucial sign error in equation 2 and some mi... more This paper has been withdrawn by the author due to a crucial sign error in equation 2 and some mistake in Table 1 information. please let me for changing this information and updating this paper.

This paper is a quick overview of Kaveh 2012 rescue virtual robot team which describes our improv... more This paper is a quick overview of Kaveh 2012 rescue virtual robot team which describes our improvements and works in relation with robotic science and AI in order to participate in Mexico 2012 Robocup Competitions. This year we are going to use some new concepts to dissolve some problems such as Multi Robot Exploration, Localization and mapping (SLAM), Victim Detection and System Architecture.

A new semi-rigid concrete beam-column connection is proposed and its performance is investigated ... more A new semi-rigid concrete beam-column connection is proposed and its performance is investigated through experimental and analytical studies. In this connection, the beam dose not directly connect to column but it is connected with torsional link member. This scissor mechanism allows the beam to make rotation with respect to column. Moreover, the strengthened connections are also suggested to support higher levels of lateral deformations for structures located in high risk seismic areas. The connections behavior under cyclic loads has been studied experimentally using three types of specimens; i) basic specimen, ii) strengthened specimen with Carbon Fiber Reinforced Polymer (CFRP) wrapping, iii) strengthened specimen with steel core reinforcing. The results showed that for low levels of deformations, the basic connection exhibited acceptable performance but for higher levels of rotation, the strengthened specimens had significant merits in ductility and nonlinear characteristics. Be...

Background detection and removal in video sequences for tracking motion objects is an important p... more Background detection and removal in video sequences for tracking motion objects is an important part of tracking process. It's clear that removal static background leads to tracking motion object and analyzing their movements more accurately. Older techniques for background removal, only use spatial features of image and remove background by focusing on color, texture …, but newer techniques improve results of background subtraction by using temporal features. In this paper we propose a novel method that uses combination of temporal and spatial features for background removal. In this method we reach to some motion windows by comparing and combining each frame with its neighbors, and obviously points out of these windows are parts of background. Then by other comparisons we detect static parts of motion windows and remove them. One advantage of this method is simplicity and the other is the ability of detection motion objects independent of their size and speed. Furthermore it has no noticeable overhead for real-time tracking process.

2014 6th Conference on Information and Knowledge Technology (IKT), 2014

ICCKE 2013, 2013

… Conference on IT to …, 2009