The simulation of vehicle counting system for traffic surveillance using Viola Jones method

Abstract

Traffic congestion in Makassar has occurred over the last 3 years due to the increase of vehicle amount with the lack of adequate road space. To solve the problem, the Intelligent Transportation System (ITS) is needed. One of the research topics in ITS is determining the turnover time of traffic lights based on the number of vehicles in the road. Viola-Jones method can be used to count the number of vehicles by detecting objects. Therefore, this study shows a simulation of the number of vehicles using the Viola-Jones method as an initial step to implement the ITS in Makassar. The data used is a front view car pictures. Simulation results show that the average accuracy rate of the detection is determined by the number of samples. For example, the average of the highest detection accuracy is 92% by using 150 positive samples and 300 negative samples were applied to 30 test samples.

Gaussian Mixture Models optimization for counting the numbers of vehicle by adjusting the Region of Interest under heavy traffic condition

“…Furthermore, previous research for vehicle counting but different method studied by Indrabayu, et al which used Viola Jones in examining accuracy effect by the number of samples. The research found highest detection accuracy was 92% [4].…”

Recognition system for facial expression by processing images with deep learning neural network

“…Traditional computer image processing programs allow transformations or beautification of digital images only, due to this, software-based applications have been devised to increase the amount of operations available to users to understand the information that an image provides. Some of these applications allow information to be recovered by reconstructing the image based on assumptions or eliminating characteristics that attenuate the information of interest [2][3][4][5].…”

“…The most common way to estimate the values of a histogram is based on: decreasing the number of dimensions of the image by converting its color format to gray scale and estimating the frequencies with the expression shown in (1). Where the dimensions are w (width) and h (height), n represents the gray levels and 𝑁 𝑛𝑔 is the number of pixels [5].…”

“…The automation of these techniques has allowed the generation of strategies and algorithms for the processing of video or sequences of images in a continuous manner. That is, the same logical structure is used to process large volumes of information, which has encouraged the development of advanced techniques such as artificial vision, closed security systems, the analysis of human gait or the location of geometric patterns [3][4][5][6][7][8]. Although there is a lot of techniques to process images, these have some limitations, because they depend on the capture device and the resolution of the image to function in an appropriate manner.…”

Sulama/İlaçlama Robotu için Nesne Tanıma Çalışmaları

“…Dewiani Djamaluddin ve arkadaşları trafik yoğunluğunu ölçebilmek için VJ nesne tanıma algoritmasından faydalanmışlardır [7]. Özellikle benzer ve simetrik içeriklere sahip taşıtların ön kısımlarından alınan az sayıdaki örnek fotoğraflar ile %92 başarılı tespit oranına erişilmiştir.…”