Arşiv logosu
  • Türkçe
  • English
  • Giriş
    Yeni kullanıcı mısınız? Kayıt için tıklayın. Şifrenizi mi unuttunuz?
Arşiv logosu
  • Koleksiyonlar
  • DSpace İçeriği
  • Analiz
  • Türkçe
  • English
  • Giriş
    Yeni kullanıcı mısınız? Kayıt için tıklayın. Şifrenizi mi unuttunuz?
  1. Ana Sayfa
  2. Yazara Göre Listele

Yazar "Ugur, Aybars" seçeneğine göre listele

Listeleniyor 1 - 1 / 1
Sayfa Başına Sonuç
Sıralama seçenekleri
  • Küçük Resim Yok
    Öğe
    Image Caption Generation Supported Information and Recommender System for Tourists, Including Supplements for Individuals With Disabilities
    (Ali KARCI, 2023) Tatar, Muhammed Salih; Kök, Rabia; Ugur, Aybars
    Tourism is one of the most important tools for individuals to interact with different cultures. A web application has been developed that includes information and recommendation system functions supported by image captioning, which will bring significant technological and social innovation to the tourism sector and includes additions for disabled individuals. In this application, a data set covering popular places in Izmir, including nearby popular, historical, and touristic places was created using location information obtained with the GPS system. The Apriori Algorithm, which is a statistical method, was used to deliver popular places more effectively to tourists. A Facilitator Module was designed that performs image captioning and object recognition functions using VGG16 and LSTM models, which are transfer learning models for user input images obtained from visually impaired individuals to help them explore the city. With this module, visually impaired individuals can describe the instant images of streets and objects in the city in textual and auditory form and can also determine whether previously entered keywords (fire, metro, statue, etc.) are present in the image. The data set created from images in MS-COCO, Flicker8k, Flicker30k and Tourism48 data sets was divided into 80% training and 20% test data. Success values were obtained as %55.41 for BLEU-1 and %30.15 for BLEU-2 in the tests conducted.

| Bursa Teknik Üniversitesi | Kütüphane | Açık Erişim Politikası | Rehber | OAI-PMH |

Bu site Creative Commons Alıntı-Gayri Ticari-Türetilemez 4.0 Uluslararası Lisansı ile korunmaktadır.


Mimar Sinan Mahallesi Mimar, Sinan Bulvarı, Eflak Caddesi, No: 177, 16310, Yıldırım, Bursa, Türkiye
İçerikte herhangi bir hata görürseniz lütfen bize bildirin

DSpace 7.6.1, Powered by İdeal DSpace

DSpace yazılımı telif hakkı © 2002-2026 LYRASIS

  • Çerez ayarları
  • Gizlilik politikası
  • Son Kullanıcı Sözleşmesi
  • Geri bildirim Gönder