Traffic Light Contoller System using Counting of Vehicles in MATLAB
We are very much aware of the fact that, the population of city and number of vehicles on the road are increasing day by day. With increasing urban population and hence the number of vehicles, need of controlling streets, highways and roads is major issue. The main reason behind today’s traffic problem is the techniques that are used for traffic management. Today’s traffic management system has no emphasis on live traffic scenario, which leads to inefficient traffic management systems. These traffic timers just show the preset time. This is like using open loop system. If we incorporate a closed loop system using camera, it is possible to predict the exact time on traffic light timers. If the traffic light timers are showing correct time to regulate the traffic, then the time wasted on unwanted green signals (green signal, when there is no traffic) will be saved. Timer for every lane is the simplest way to control traffic. And if those timers are predicting exact time then automatically the system will be more efficient. This paper represents the project that has been implemented by using the Matlab software and it aims to prevent heavy traffic congestion. This project measure the number of vehicles present on the road. At first, film of highway is captured by a camera has been installed in highway. A web camera is placed in a traffic lane that will capture images of the road on which we want to control traffic. Then these images are efficiently processed to know the traffic density. According to the processed data from matlab, the controller will send the command to the timer to show particular time on the signal to manage traffic.
PROJECT OUTPUT VIDEO:
Chandrasekhar. M, Saikrishna.C, Phaneendra Kumar propose the implementation of image processing algorithm in real time traffic light control which will control the traffic light efficiently. A web camera is placed in each stage of traffic light that will capture the still images of the road where we want to control the traffic. Then those captured images are successively matched using image matching with a reference image which is an empty road image. The traffic is governed according to percentage of matching.
The key point of the paper is the technique which is used for image comparison. The authors have used image matching technique. SIFT algorithm is been used in this paper and this is very effective and pretty simple.
Vikramaditya Dangi, Amol Parab, Kshitij Pawar & S.S Rathod propose the way to implement an intelligent traffic controller using real time image processing. The image sequences from a camera are analyzed using various edge detection and object counting methods to obtain the most efficient technique. Subsequently, the number of vehicles at the intersection is evaluated and traffic is efficiently managed. The paper also proposes to implement a real-time emergency vehicle detection system. In case an emergency vehicle is detected, the lane is given priority over all the others.
The key point of this paper is the technique which is used for edge detection. The authors have given the comparison of various edge detection techniques and conclude that canny edge detection is the best method for edge detection. Thus we are using canny edge detection.
DISADVANTAGES OF EXISTING SYSTEM:
Present Traffic Light Controllers are based on microcontrollers & microprocessors. These TLC have limitations because it uses the pre-defined hardware, which is functioning according to the program that does not have the flexibility of modification on real time basis.
In this paper, a real-time traffic congestion estimation approach was proposed, which is based on image texture feature extraction and texture analysis. The main innovations are as follows.
First, proposing human- computer interaction approach to set vehicle area. Which is not only faster than the common used background training method, but more convenient to select area of interesting.
Second, we proposed extracting texture features to estimate vehicle density. Researchers have used texture analysis method to estimate the density of pedestrian and fish, but no one else used it to estimate the density of vehicle.
Our experimental results showed using our new texture feature to estimate vehicle density, the accuracy could be as high as 99%, and the speed is very fast, which could meet the real-time demand in engineering.
We believe that our approach would be integrated in road video surveillance system in the future and provide reliable and fast road information for traffic managers.
ADVANTAGES OF PROPOSED SYSTEM:
In this paper, a real-time road congestion detection algorithm based on texture analysis is proposed, which deals with image data from road surveillance systems and carries out the accurate identification of vehicle density in different scenes.
It is considered to successfully provide quick and reliable traffic information to the traffic administrative departments.
- Vehicle Area calibration
- GLCM calculation
- Road congestion recognition
Vehicle Area calibration
The current method of congestion detection basically carried out background training at first. However the proceeding is not only time-consuming, but easily affected by many factors. For time-saving, the paper proposed a fast method of human-computer interaction. That is artificially setting vehicle area firstly, and then using the texture analysis method for congestion estimation. After the setting, the gray value of vehicle area remains the same, while the other area is set to zero, which means the background. This approach is time-saving, at the same time the interesting area can be selected freely.
The standard gray image is 256 gray scale, the corresponding GLCM is 256X256. Computing GLCM with this size is both time-consuming and not necessary. It is well known that mostly the colour of vehicle is single, so in image the gray value of vehicle should be single or several successive values. But because of the light reflection, actually the gray value of vehicle area may have dozen choices. It results in that GLCM histogram distribution is more homogeneous than real, therefore the extracted texture features cannot truly reflect the density of vehicles. To solve this problem, we can reduce the gray levels. After testing, it is suitable to reduce 256-level to 32-level. In 32-level image, the pixel value in the area of black vehicle only has less than 3 choices.
The GLCM can be calculated after gray scale reduction. In our approach, d is set to 1, that is to calculate the distribution of adjacent pixels. This method can be applied to deal with images taken from different places. As mentioned before, after gray scale reducing the area with zero pixel value was treated as background. When carrying out texture analysis on the image, the background area should be abandoned. So, the first row and the first column of GLCM are removed. Consequently, the size of GLCM is changed from 32X32 to 31X31.
Road congestion recognition
Feature S reflects the density of vehicle in the image, bigger S means heavier road congestion, smaller S means unobstructed road condition. We carried out hundreds of experiments to obtain the decision threshold ST. When S>ST, there are crowded vehicles on the road, which may lead to traffic congestion. When S<ST , the traffic is smooth. In the next section, the detail of obtaining threshold ST will be introduced
System : Pentium Dual Core.
Hard Disk : 120 GB.
Monitor : 15’’ LED
Input Devices : Keyboard, Mouse
Ram : 1GB.
Operating system : Windows 7.
Coding Language : MATLAB.
Tool : MATLAB R2013A /2018
Li Weia, Dai Hong-ying, “Real-time Road Congestion Detection Based on Image Texture Analysis”, Procedia Engineering 137 ( 2016 ) 196 – 201.