Views
22
Downloands
2
Günümüzde teknolojinin hızla ilerlemesi ile birlikte yapay zeka hayatımızın vazgeçilmez bir parçası haline gelmiştir. Yapay zeka, bilgisayar ya da bilgisayar ile kontrol edilen faaliyetlerin insan ya da zeki canlılar gibi tahmin edebilme yöntemi olarak tanımlanabilmektedir. Yapay zeka, sağlık, eğitim, güvenlik, robotic gibi birçok uygulama alanında sıklıkla kullanılmaktadır. Yapay zekanın önemli kullanım alanlarından birisi de araçların kontrollü ve güvenli bir biçimde geçişleri için kullanılan trafik sinyalizasyon sistemleridir. Trafik sinyalizasyonu genellikle trafiğin yoğun olduğu kavşaklarda trafiğin yoğunluğuna bakmaksızın sabit süre bazlı çalışmaktadır. Bu nedenle sabit süreli trafik sinyalizasyon sistemleri günümüzde çok tercih edilmemektedir. Akıllı trafik sinyalizasyon sistemlerinde bekleme süresi ise araç sayısı ve araçların geçiş sürelerine bağlıdır. Çalışmada Isparta ilinde bulunan Ulaştırma ve Trafik Hizmetleri Müdürlüğü’nden özellikle trafiğin yoğun olduğu kavşaklardan birisi olan Otogar kavşağındaki araçlara ait video görüntüleri alınmıştır. Alınan görüntüler üzerinde görüntü işleme ve ARIMA derin öğrenme yöntemi uygulanmıştır. ARIMA derin öğrenme yöntemi ile zamana bağlı araç sayısı ve araç geçiş süreleri R2 performans değerlendirme ölçütüne göre değerlendirilerek %82 ve %89 doğruluk oranı elde edilmiştir.
Nowadays, with the rapid advancement of technology, artificial intelligence has become an integral part of our life. Artificial intelligence can be defined as a method for predicting activities controlled by a computer or computer, such as humans or intelligent beings. Artificial intelligence is frequently used in many applications such as health, education, safety, robotics. One of the important areas of use of artificial intelligence is traffic signaling systems used for the controlled and secure passage of vehicles. Traffic signaling is usually based on fixed time, regardless of the traffic intensity in the crossroads where traffic is intense. Therefore, fixed-term traffic signaling systems are not much preferred today. The waiting time in the smart traffic signaling systems depends on the number of vehicles and the time of passage of the vehicles. In the study, video images of the vehicles in the Autogar cross, which is one of the crossroads where traffic is especially intense, were taken from the Transport and Traffic Services Directorate located in the Isparta province. Image processing and the ARIMA deep learning method is applied to the received images. With the ARIMA deep learning method, the number of time-related vehicles and passing times of vehicles were assessed according to the performance assessment measure R2 and the accuracy rate was 82% and 89% achieved.
Today, with the rapid advancement of technology, artificial intelligence has become an indispensable part of ourlives. Artificial intelligence can be described as a method of predicting computer or computer-controlled activities like human or intelligent creatures. Artificial intelligence are frequently used in many application areas such as health, education, security and robotics. One of the important uses of artificial intelligence is traffic signaling systems used for the controlled and safe passage of vehicles. Traffic signaling generally works on fixed time basis at intersections with heavy traffic, regardless of the traffic density. Thus, fixed time traffic signaling systems are not preferred much today. The waiting time in smart-traffic signaling systems depends on the number of vehicles and the transit time of the vehicles. The study, video footage of the vehicles at the Otogar intersection, which is one of the intersections with heavy traffic, was taken from the Directorate of Transport and Traffic Services in Isparta. Image processing and ARIMA deep-learning method were applied on the captured images. With the ARIMA deep-learning method, the number of time-dependent vehicles and the R2 of vehicle transit times were evaluated according to the performance evaluation criteria and accuracy rate of 82% and 89% was obtained.
Field : Mimarlık, Planlama ve Tasarım; Mühendislik
Journal Type : Uluslararası
Article Detail 
Similar Articles
Download PDF
Journal Information
Listen Paper
Cite
Print This Page
Share
Mendeley
