Prototyping a Traffic Light Recognition Device with Expert Knowledge

Traffic light detection and recognition (TLR) research has grown every year. In addition, Machine Learning (ML) has been largely used not only in traffic light research but in every field where it is useful and possible to generalize data and automatize human behavior. ML algorithms require a large amount of data to work properly and, thus, a lot of computational power is required to analyze the data. We argue that expert knowledge should be used to decrease the burden of collecting a huge amount of data for ML tasks. In this paper, we show how such kind of knowledge was used to reduce the amount of data and improve the accuracy rate for traffic light detection and recognition. Results show an improvement in the accuracy rate around 15%. The paper also proposes a TLR device prototype using both camera and processing unit of a smartphone which can be used as a driver assistance. To validate such layout prototype, a dataset was built and used to test an ML model based on adaptive background suppression filter (AdaBSF) and Support Vector Machines (SVMs). Results show 100% precision rate and recall of 65%.

Download Full-text

Frequency Maps as Expert Instructions to Lessen Data Dependency on Real-time Traffic Light Recognition

10.20944/preprints202008.0138.v1 ◽

2020 ◽

Author(s):

Hendrik Macedo ◽

Thiago Almeida ◽

Leonardo Matos ◽

Bruno Prado

Keyword(s):

Real Time ◽

Autonomous Vehicles ◽

Machine Learning Algorithms ◽

Support Vector ◽

Background Suppression ◽

Processing Unit ◽

Traffic Light ◽

Traffic Lights ◽

Real Time Traffic ◽

Computational Resources

Research on Traffic Light Recognition (TLR) has grown in recent years, primarily driven by the growing interest in autonomous vehicles development. Machine Learning algorithms have been widely used to that purpose. Mainstream approaches, however, require large amount of data to properly work, and as a consequence, a lot of computational resources. In this paper we propose the use of Expert Instruction (IE) as a mechanism to reduce the amount of data required to provide accurate ML models for TLR. Given an image of the exterior scene taken from the inside of the vehicle, we stand the hypothesis that the picture of a traffic light is more likely to appear in the central and upper regions of the image. Frequency Maps of traffic light location were thus constructed to confirm this hypothesis. The frequency maps are the result of a manual effort of human experts in annotating each image with the coordinates of the region where the traffic light appears. Results show that EI increased the accuracy obtained by the classification algorithm in two different image datasets by at least 15%. Evaluation rates achieved by the inclusion of EI were also higher in further experiments, including traffic light detection followed by classification by the trained algorithm. The inclusion of EI in the PCANet achieved a precision of 83% and recall of 73% against 75.3% and 51.1%, respectively, of its counterpart. We finally presents a prototype of a TLR Device with that expert model embedded to assist drivers. The TLR uses a smartphone as a camera and processing unit. To show the feasibility of the apparatus, a dataset was obtained in real time usage and tested in an Adaptive Background Suppression Filter (AdaBSF) and Support Vector Machines (SVMs) algorithm to detect and recognize traffic lights. Results show precision of 100% and recall of 65%.

Download Full-text

Comparative Performance of Machine Learning Algorithms for Cryptocurrency Forecasting

Indonesian Journal of Electrical Engineering and Computer Science ◽

10.11591/ijeecs.v11.i3.pp1121-1128 ◽

2018 ◽

Vol 11 (3) ◽

pp. 1121 ◽

Cited By ~ 4

Author(s):

Nor Azizah Hitam ◽

Amelia Ritahani Ismail

Keyword(s):

Machine Learning ◽

Time Series Data ◽

Learning Algorithms ◽

Machine Learning Algorithms ◽

Series Data ◽

Support Vector ◽

Small Range ◽

Accuracy Rate ◽

Comparative Performance ◽

Vector Machines

Machine Learning is part of Artificial Intelligence that has the ability to make future forecastings based on the previous experience. Methods has been proposed to construct models including machine learning algorithms such as Neural Networks (NN), Support Vector Machines (SVM) and Deep Learning. This paper presents a comparative performance of Machine Learning algorithms for cryptocurrency forecasting. Specifically, this paper concentrates on forecasting of time series data. SVM has several advantages over the other models in forecasting, and previous research revealed that SVM provides a result that is almost or close to actual result yet also improve the accuracy of the result itself. However, recent research has showed that due to small range of samples and data manipulation by inadequate evidence and professional analyzers, overall status and accuracy rate of the forecasting needs to be improved in further studies. Thus, advanced research on the accuracy rate of the forecasted price has to be done.

Download Full-text

Road-Sign Detection and Recognition Based on Support Vector Machines

IEEE Transactions on Intelligent Transportation Systems ◽

10.1109/tits.2007.895311 ◽

2007 ◽

Vol 8 (2) ◽

pp. 264-278 ◽

Cited By ~ 392

Author(s):

Saturnino Maldonado-Bascon ◽

Sergio Lafuente-Arroyo ◽

Pedro Gil-Jimenez ◽

Hilario Gomez-Moreno ◽

Francisco Lopez-Ferreras

Keyword(s):

Support Vector Machines ◽

Support Vector ◽

Road Sign ◽

Vector Machines ◽

Sign Detection ◽

Detection And Recognition

Download Full-text

Traffic light detection and recognition based on Haar-like features

10.23919/elinfocom.2018.8330598 ◽

2018 ◽

Cited By ~ 3

Author(s):

Sang-Hyuk Lee ◽

Jung-Hawn Kim ◽

Yong-Jin Lim ◽

Joonhong Lim

Keyword(s):

Traffic Light ◽

Light Detection ◽

Detection And Recognition

Download Full-text

Sentiment Analysis towards Actionable Intelligence via Deep Learning

TEM Journal ◽

10.18421/tem94-44 ◽

2020 ◽

pp. 1663-1668

Author(s):

Shorouq Fathi Eletter

Keyword(s):

Deep Learning ◽

Sentiment Analysis ◽

Service Providers ◽

Unstructured Data ◽

Classification Model ◽

Online Review ◽

Support Vector ◽

Competitive Advantages ◽

Accuracy Rate ◽

Vector Machines

The exponential growth of unstructured data and the ability of businesses to utilize such data in decision-making have led to competitive advantages. The knowledge provided by analyzing unstructured data is crucial for product developers or service providers because it might affect the sustainability of the business. Sentiment analysis is used to gain an understanding of the attitudes, opinions, and emotions expressed within an online review. Naïve Bayes (NB), logistic regression (LR), decision trees (DT), deep learning (DL), and support vector machines (SVM) were used to build a classification model. In the data mining settings, the classification accuracy is the best metric to highlight the best classifier. The DL classifier outperformed other models in terms of accuracy rate. Classifying customers' feelings toward a product or service is critical for providing actionable insights. Utilizing such models will help to analyze huge volumes of reviews, saving both time and costs.

Download Full-text

DETEKSI DAN REKOGNISI RAMBU-RAMBU LALU LINTAS DENGAN MENGGUNAKAN METODE SUPPORT VECTOR MACHINE

Jurnal Ilmiah Teknologi Infomasi Terapan ◽

10.33197/jitter.vol3.iss2.2017.131 ◽

2017 ◽

Vol 3 (2) ◽

Author(s):

Kurniawan Nur Ramadhani ◽

M.Syahrul Mubarok ◽

Agnes Dirgahayu Palit

Keyword(s):

Support Vector Machines ◽

Road Safety ◽

Support Vector ◽

Traffic Sign ◽

Traffic Signs ◽

Histogram Of Oriented Gradient ◽

Vector Machines ◽

Sign Detection ◽

Traffic Sign Detection ◽

Detection And Recognition

[Id]Kota-kota besar pasti tidak lepas dengan penggunaan rambu lalu lintas untuk meningkatkan keselamatan pengguna jalan. Rambu lalu lintas dirancang untuk pembantu pengemudi untuk mencapai tujuan mereka dengan aman, dengan menyediakan informasi rambu yang berguna. Meskipun demikian, hal yang tidak diinginkan dapat terjadi ketika informasi yang tersimpan pada rambu lalu lintas tidak diterima dengan baik pada pengguna jalan. Hal ini dapat menjadi masalah baru dalam keamanan berkendara. Dalam meminimalisasi masalah tersebut, dapat dibuat suatu teknologi yang mengembangkan sistem yang mengidentifikasi objek rambu lalu lintas secara otomatis yang dapat menjadi salah satu alternatif meningkatkan keselamatan berkendara, yaitu Traffic Sign Detection and Recognition (Sistem Deteksi dan Rekognisi Rambu Lalu Lintas). Sistem ini menggunakan menggunakan deteksi ciri warna dan bentuk. metode Histogram of Oriented Gradient (HOG) untuk ektraksi ciri citra bentuk, colour moment untuk ekstraksi warna dan Support Vector Machines (SVM) untuk mengklasifikasikan citra rambu lalu lintas. Sehingga dapat dianalisa bagaimana Sistem dapat mendeteksi dan mengenali citra yang merupakan objek rambu lalu lintas Diharapkan dengan adanya paduan metode-metode tersebut dapat membangun sistem deteksi dan rekognisi rambu lalu lintas, dan meningkat performansi sistem dalam mendeteksi dan mengenali rambu lalu lintas. Performansi yang dihasilkan dari sistem adalah 94.5946% menggunakan micro average f1-score.Kata kunci : ekstraksi ciri fitur, ekstraksi ciri warna, klasifikasi, HOG, colour moment, SVM, micro average f1-score.[En]The big cities must not be separated by the use of traffic signs to improve road safety. Traffic signs are designed to aide drivers to reach their destination safely, by providing useful information signs. Nonetheless, undesirable things can happen when information stored in the traffic signs are not received well on the road. It can be a new problem in road safety. In minimizing the problem, can be made of a technology that is developing a system that identifies an object traffic signs automatically which can be one alternative to improve driving safety, the Traffic Sign Detection and Recognition (Detection System and Traffic Sign Recognition). The system uses using the detection characteristics of colors and shapes. methods Histogram of Oriented Gradient (HOG) to extract image characteristic shape, color moment for the extraction of color and Support Vector Machines (SVM) to classify traffic signs image. So it can be analyzed how the system can detect and recognize the image which is the object of traffic signs Expected by the blend of these methods can build a system of detection and recognition of traffic signs, and increased system performance to detect and recognize traffic signs. Performasi generated in the system is 94.5946% using micro average f1-score.

Download Full-text