ارسل ملاحظاتك

ارسل ملاحظاتك لنا







الرئيسية > Image Feature Matching for Vis...>الوصف

Image Feature Matching for Visual Object Recognition

المؤلف الرئيسي: Al Qaisi, Taghreed (Author)
مؤلفين آخرين: Olszewska, Joanna Isabelle (Advisor)
التاريخ الميلادي: 2012
موقع: هدرسفيلد
الصفحات: 1 - 61
رقم MD: 752088
نوع المحتوى: رسائل جامعية
اللغة: الإنجليزية
الدرجة العلمية: رسالة ماجستير
الجامعة: University of Huddersfield
الكلية: School of Computing and Engineering
الدولة: بريطانيا
قواعد المعلومات: +Dissertations
مواضيع:
التعرف على الأشياء
إدارة المعلومات البصرية
رابط المحتوى:
صفحة العنوان     
المستخلص     
قائمة المحتويات     
24 صفحة الأولى     

الناشر لهذه المادة لم يسمح بإتاحتها.
صورة الغلاف QR قانون
حفظ في:
المستخلص: Due to the wide use of visual information and the rapid development of technologies in this area, recent research in computer vision has involved the study of visual object recognition, and many object recognition algorithms have been introduced. However, the object recognition subject is still considered as a challenging task, since similarity measuring between two images is a problem in itself. This thesis offers an automatic object recognition approach for visual information management. In this work, the method we have developed has two main stages, namely, feature extraction and feature matching. As feature extraction method, we have adopted the Scale Invariant Feature Transform (SIFT) detector, while our feature matching method is based on the Hausdorff distance. Our approach has been implemented and evaluated as an object category recognition system. The experiments for both feature extraction (detection) and feature matching (recognition) stages have demonstrated that the project has achieved its goal with good performance. Furthermore, some changes in the Hausdorff distance algorithm using the City-Block distance instead of Euclidean distance have brought improvements in terms of computational time saving.

عناصر مشابهة


© 2024 المنظومة. جميع الحقوق محفوظة.