Improving Motion Detection Using a Hybrid Method

الموضوعات

علوم تطبيقية وتكنولوجية

الملخص

Motion detection is the action of sensing the physical movement in a given area. It is the process of determining the movement of an object from two or more successive images, by labelling moving objects shot in front of a background that has little or no movement. This is typically the case for applications using a fixed camera such as video surveillance applications or traffic monitoring applications. This thesis attempts to propose a system for motion detection based on hybrid methods which combine (background subtraction technique based on approximated median filter) and (shadow detection and removal technique based on color space using the hybrid color space(HSV and YCbCr)), under dynamic background with static camera. The proposed system was tested using indoor and outdoor scenes with humans, vehicles, and motorcycles in different times. These test image sequences included various weather and lighting conditions, including periods of sunny ,sunset, rainy, noisy and darkness and different light source. The results show system stability in extraction of background and object in all previous situations which tested for it. The results offer processing speed for background modeling algorithm approximated median filter (AMF) is as high as a frame rate of 15 frames per second this leads to speed up to 50% than the visual background extractor (ViBe) algorithm which is give processing speed as high as a frame rate of 10 frames per second. Furthermore, the proposed system gives error rate (0.042%) of the foreground segmentation and the background extraction as compared with the average error rate of mixture of Gaussain algorithm which was 0.52% . This means that the proposed system offers less error rate, stable and can continuously produce a better quality for background extraction and object segmentation.

روابط وملفات

التعريف والنوع

رقم الوثيقة

0f4187b5-b9c6-4565-8413-204f11a7727b

رقم العقد

0

نوع الوسائط

Crawler

نوع المحتوى

الرسائل العلمية

صيغة المصدر

رسائل ماجيستير

نوع الملف

pdf image

أسماء الملفات

0f4187b5-b9c6-4565-8413-204f11a7727b_1.pdf

بيانات النشر

ترجمة العنوان

تحسين اكتشاف الحركة باستخدام طريقة هجينة

ألقاب المؤلفين

[{"name_ar":"زينب محمد حسين","title_ar":"اشراف","title_en":"Supervision"},{"name_ar":"علي أطعيمة جبر","title_ar":"اعداد","title_en":"Preparation"}]

اللغة

English

المصدر والدورية

اسم المصدر

Improving Motion Detection Using a Hybrid Method

المحتوى والصفحات

عدد الصفحات

0

إشراف وإعداد

الإشراف

زينب محمد حسين

الإعداد

علي أطعيمة جبر

الاقتباسات الببليوغرافية