Author: Hazem Noori Abdulrazzak

Neural Network (NN) is an information processing system that has been developed as generalization models of human cognition of neural biology. Two important learning methods for NN are discussed to be our comparative study; these methods are Back Propagation (BP) and Particle Swarm Optimization (PSO). These two methods are found to learn the NN by modifying the weights of the NN, this is done by calculating the fitness value which is considered as a threshold value. In this work the Pattern Recognition (PR) problem will be manipulated, which is considered one of the important applications of classification filed. The Greek Text Letter Recognition (A) is the objective of this paper, the upper and lower cases are discussed. DOI LINK: link

In the world today, traffic congestion in most cities is a severe problem. Recently, the system used is based on a fixed-time system in traffic, which cannot provide incompetence if one or more lane has higher traffic than others. The higher traffic on one side lane needs a longer operation time for a green signal as compared to the fixed time-based systems. This paper proposed a High Priority Traffic Light system HPTL as a technique to compute the period time of green and red lights based on the density of traffic present at that time. To fix this problem an automatic traffic light control system should proposed. The proposed system will be achieved by using PIR  proximity Infrared Sensors. When the density is computed, the luminous period of the green and red signal will be selected to help the microcontroller. On each roadside, the sensors can detect the traffic density and calculate the vehicles in a specific lane then forward the density value to the microcontroller. The microcontroller will select the free lane and the high-traffic lane and then assign the priority system. The proposed system is successful in investing time, reducing vehicle density, and increasing reliability.