Traffic light control utilizing queue length and object segmentation

العناوين الأخرى

التحكم بالإشارة الضوئية من خلال استعمال طول الطابور و اكتشاف الأجسام

مقدم أطروحة جامعية

al-Ayish, Ubadah Muhammad Adnan

مشرف أطروحة جامعية

Samawi, Venus W.

أعضاء اللجنة

al-Smadi, Adnan M.
Slait, Azzam Talal
Hamarishah, Akram

الجامعة

جامعة آل البيت

الكلية

كلية الأمير الحسين بن عبد الله لتكنولوجيا المعلومات

القسم الأكاديمي

قسم علوم الحاسوب

دولة الجامعة

الأردن

الدرجة العلمية

ماجستير

تاريخ الدرجة العلمية

2010

الملخص الإنجليزي

As the social economic activities grow variously in recent days, an advanced traffic control system corresponding to the changes of road traffic circumstances becomes an urgent matter.

This research suggests two techniques based on picture taken by digital camera placed at fixed position in the traffic light with specific resolution and distance.

The resulted image will be preprocessed (subtract image from its background, remove noisy small objects) then use two techniques queue length (find out the length of car queue on road depending on the distance between the two end points on road line), and vehicle counting (by performing image segmentation and count out segments that satisfy the proper size of vehicle).To measure the queue length or to count the number of the vehicles, edge detection (in this research Sobol edge detection method is used) and segmentation (based on region growing) are needed.

Finally, an equation to find out the estimated time and actual time is suggested to determine the estimated time reference for optic Green.

The two suggested techniques compared from two points of views: time needed to make decision of green-light period, and the suggested estimated time by the technique with respect to the actual time needed.

As a result, it was found the vehicle counting technique is recommended to be used since it shows better performance from decision making time and estimated time with respect to actual.

Also the Impact of image quality (low and high density of vehicles on road) is studied, where it was found that the queue suit high density road vehicle counting suit low density road, but generally segmentation suitable for both types.

التخصصات الرئيسية

تكنولوجيا المعلومات وعلم الحاسوب

الموضوعات

• أنظمة التحكم
• إشارات الطرق

عدد الصفحات

63

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-314908