Traffic Image Processing Essay

Intelligent Traffic Light Control Using Image Processing

Rahul Yadav1, Rahul Gupta2, Aditi Versha3, Vishal Yadav4

1,2,3,4 Department of Electronics and Telecommunication, Lokmanya Tilak college of Engineering, Mumbai University

1 [e-mail],2 [e-mail],[e-mail],[e-mail]

Abstract— Traffic congestion at junctions has become a familiar issue. The density of vehicles is increasing day by day on the road and there is a requirement of adaptive traffic signals which can do real time monitoring of traffic density. The aim is to develop a system that efficiently uses image processing to tackle the heavy traffic on the roads. A step by step approach of image acquisition, image processing

METHODOLOGY

This paper discusses and addresses the issue of recognizing and tracking the vehicle using Color blob detection technique [1]. Firstly it is essential to differentiate the foreground from background. This is done using a foreground detector. A foreground detector differentiates between the foreground and the background by learning about the frames in the background. An object is detected, and it is enclosed in a rectangular space. Morphological operations have been applied to even out the irregularities/ inconsistency. The final counting is done by tracking the detected objects and regions.

The procedure adopted to monitor the traffic density is divided into three phases. The first phase deals with the image acquisition. The second phase handles the object detection using color blob analysis. The third phase is the image matching that is done between the reference image and the captured image. The result of this matching will determine the duration of the green signal assigned to each

IMAGE ACQUISITION

A web camera equipped with a stepper motor is installed on the road junction. Stepper motor is responsible for capturing images from all directions in steps of fixed time interval. The speed of rotation of the camera is designed to be such that it is greater than the click-to-capture time of the camera. The image is captured by a webcam and it is then transferred to computer via a USB cable. The image acquisition and processing is done by using the image acquisition toolbox of MATLAB.

V. IMAGE PROCESSING Color extraction: The image captured is a color image as shown in fig2.1. It has 3 components: R (red), G (green) and B (blue). To obtain accuracy in the image processing the RGB image is converted to HSV image (hue, saturation, value). After converting the image, the hue, saturation and value components are extracted and threshold is applied to each colour band.

VI. THRESHOLDING

Thresholding is process of converting grayscale image to a bi-level image using an optimal threshold. The purpose of Thresholding is to extract those pixels from the image which represent an object. Though the information is binary the pixels represent range of intensities. Thus the objective of binarization is to mark the pixels that belong to foreground regions with a single intensity and background regions with different intensity. The hue and saturation extracted image are threshold to convert this image into binary