Türk halılarının görüntü veri tabanı kullanarak saklanması ve sorgulanması
Başlık çevirisi mevcut değil.
- Tez No: 75158
- Danışmanlar: DOÇ. DR. MUHİTTİN GÖKMEN
- Tez Türü: Yüksek Lisans
- Konular: Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Computer Engineering and Computer Science and Control
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1998
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Bilgisayar Mühendisliği Ana Bilim Dalı
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 154
Özet
ÖZET Görüntü, bir veri tipi olarak çok yer kaplayan, yönetimi zor, klasik veritabanı yönetim sistemlerinde sakladığı zaman çeşitli sorunlara yer açan bir yapıdadır. Bu nedenle verilerin saklanması ve sorgulanması konusunda çeşitli alternatif yaklaşımlar izlenmektedir. Bu alternatif yaklaşımlardan biri görüntü tipi verileri dosya sistemi içinde saklamak, sadece sorgulama için gereken bilgileri bir veritabanı yönetim sistemi içinde tutmaktır. Bu veritabanı sistemi ilişkisel tabanlı bir SQL sunucusudur. Burada markaya bağlı kalmamak için ANSI SQL kullanmak faydalı olacaktır. Orta katmanda geliştirilen nesneler veriye ulaşmak için kullanılmakta ve bir tip kütüphanesi oluşturmaktadır. Bu tip kütüphanesinin en önemli üç öğesi HALI, DESEN ve SORGULAMAdır. HALI nesnesi halı üzerindeki işlemler için bir kütüphane oluştururken, DESEN nesnesi aynı görevi desenler için yerine getirir. SOGULAMA nesnesi sorgulama sonuçlarını bir collection(kolleksiyon) nesnesine yerleştirir. Ardından bu collection nesnesi kullanıcı arabiriminde gösterilecek halılar için taban oluşturur. Kullanıcı arayüzünde gerçeklenen formlar kullanıcının halıları işleyebilmesini, desenleri tanımlayabilmesini ve çeşitli değişiklikler yapabilmesini sağlamaktadır.
Özet (Çeviri)
SUMMARY Storing and Querying Turkish Carpets Using an Image Database Images are being generated at an ever-increasing rate by sources such as defense and civilian satellites, military reconnaissance and surveillance flights, fingerprinting and mug-shot-capturing devices, scientific experiments, biomedical imaging, and home entertainment systems. For example, NASA's Earth Observing System will generate about 1 terabyte of image data per day when fully operational. A content-based image retrieval (CBIR) system is required to effectively and efficiently use information from these image repositories. Such a system helps users (even those unfamiliar with the database) retrieve relevant images based on their contents. Application areas in which CBIR is a principal activity are numerous and diverse: art galleries and museum management, architectural and engineering design, interior design, remote sensing and management of earth resources, geographic information systems, scientific database management, weather forecasting, retailing, fabric and fashion design, trademark and copyright database management, law enforcement and criminal investigation, and picture archiving and communication systems. With the recent interest in multimedia systems, CBIR has attracted the attention of researchers across several disciplines. Previous approaches Previous approaches to content-based retrieval have taken two directions. In the first, image contents are modeled as a set of attributes extracted manually and managed within the framework of conventional database-management systems. Queries are specified using these attributes. Attribute-based representation of images entails a high level of image abstraction. Generally, the higher the level of abstraction, the lesser is the scope for posing ad hoc queries to the image database. Attribute-based retrieval is advocated and advanced primarily by database researchers. XIThe second approach depends on an integrated feature-extraction/object-recognition subsystem to overcome the limitations of attribute-based retrieval. This subsystem automates the feature-extraction and object-recognition task that occurs when the image is inserted into the database. However, automated approaches to object recognition are computationally expensive, difficult, and tend to be domain specific. This approach is advanced primarily by image-interpretation researchers. Recent research Recent CBIR research recognizes the need for synergy between these two approaches. Toward this goal, efforts draw upon ideas from areas such as knowledge-based systems, cognitive science, user modeling, computer graphics, image processing, pattern recognition, database-management systems, and information retrieval. This confluence of ideas has culminated in the introduction of novel image representations and data models, efficient and robust query-processing algorithms, intelligent query interfaces, and domain-independent system architectures. These advances have brought CBIR systems from their infancy to a state of reasonable maturity. Primitive versus logical Current approaches to CBIR differ in terms of which image features are extracted, the level of abstraction manifested in the features, and the degree of desired domain independence. There are two major categories of features: primitive and logical. Primitive, or low-level, image features such as object centroids and boundaries can be extracted automatically or semiautomatically. Logical features are abstract representations of images at various levels of detail. Some logical features may be synthesized from primitive features whereas others can only be obtained through considerable human involvement. Logical features denote the deeper domain semantics manifested in the images. Trade-offs In developing CBIR systems, there is an inherent trade-off between the degree of automation desired for feature extraction and the level of domain independence realized in the system. CBIR systems can be developed with emphasis on automatic and dynamic feature extraction. Although some features may be determined a priori, these systems can dynamically compute the required primitive features and synthesize the logical ones, both under the guidance of a domain expert. XllThis approach is ambitious and aims at sophisticated CBIR; systems employing mis approach are most suitable for applications involving relatively small image collections and when retrieval is performed exclusively by domain experts. We call this the dynamic feature-extraction approach. CBIR systems can also be developed that achieve a reasonable degree of domain independence at the cost of not having a completely automated system for feature extraction. We refer to this approach as a priori feature extraction. A set of primitive features is extracted, and all logical features are derived only when the image is inserted into the database. Primitive features are typically derived semiautomatically, while the logical features are extracted manually or semiautomatically. Queries are processed using both primitive and logical features. Query classes Regardless of which approach is used, generic query classes5 facilitate CBIR through retrieving by color, texture, sketch, shape, volume, spatial constraints, browsing, objective attributes, subjective attributes, motion, text, and domain concepts. An image retrieval system featuring all these query classes will have reasonable generality for dealing with diverse applications. Color and texture queries let users select images containing objects specified accordingly. The notion of an image object is domain dependent and represents a semantic entity of interest in an application. Retrieval by sketch lets users outline an image and then retrieves a like image from the database. This class can be thought of as retrieving images by matching the dominant edges. The shape class of queries has a counterpart in 3D images referred to as Retrieval by Volume. The spatial constraints category deals with a class of queries based on spatial and topological relationships among the objects in an image. These relationships may span a broad spectrum ranging from directional relationships to adjacency, overlap, and containment involving a pair of objects or multiple objects. XlllRetrieval by browsing is performed when users are vague about their retrieval needs or are unfamiliar with the structure and types of information available in the image database. The objective attributes query uses attributes like the date of image acquisition or the number of bedrooms in a residential floor-plan image and is similar to Structured Query Language retrieval in conventional databases. Retrieval is based on an exact match of attribute values. In contrast, a subjective attributes query is characterized by the presence of attributes that may be interpreted differently by each user. For example, in a mug-shot database, the attribute eyebrow shape assumes one of three values: arched, normal, or straight. One user may assign the normal value for the eyebrow shape, while another may interpret the value as arched. Retrieval by motion facilitates retrieving spatiotemporal image sequences depicting a domain phenomenon that varies in time or geographic space. Some applications require retrieving images based on associated text. Such a need is modeled by retrieval by text. Note that processing this query involves natural language processing and information retrieval techniques. The above query classes can be used as fundamental operators in formulating a class of complex queries referred to as Retrieval by Domain Concepts. An example of this is“Retrieve images of snow-covered mountains.”Not all the above generic query classes are necessary, however, for a given image retrieval application. For example, a real estate marketing application may require only retrieval by browsing, objective attributes, shape, and spatial constraints. In such a case, the a priori feature-extraction approach helps generate an application- specific system from a generic one by retaining only the necessary query classes. Though this approach doesn't entail the level of sophistication of the dynamic feature-extraction approach, it appears to be the most promising and realistic. Since users are not expected to be familiar with the image-interpretation task, the a priori feature-extraction approach has broader appeal and suits naive and casual users. It should be emphasized that true CBIR can only be achieved by synergistically employing various generic query classes in a way transparent to the user. In other words, Retrieval by Domain Concepts queries need to be expressed as a composition of the fundamental operators in a domain-independent way without involving the user. Furthermore, images don't occur in isolation; CBIR issues must be addressed in a broader information retrieval context. For example, diagnostic medical images are retrieved not only in terms of image contents but also in terms of other information associated with the images (like text describing a physician's diagnosis, treatment plan, and the final outcome). Hence, from the physician's viewpoint, the text associated with diagnostic medical images is as central as the content of the image itself. Therefore, information retrieval techniques5 have an important complementary role in CBIR. XIVProject details On that project carpets will be saved on the file system. Relational Database will be used for retrieving, querying and saving the properties (field names). On the middle tier there will be objects support database operaions. The basic clasess are designed HALI (carpet), DESEN (Figure, motif) and SORGULAMA (Query). HALI manages operations on carpets, DESEN manages operations on figures. SORGULAMA creates queries according to the info comes from query form. There are a few user interface forms developed. Using that forms customer can easily enter carpet and motif info to the database. After that using various query forms he can make queries and get result on a collection object. Every collection object is the base for a new carpet set.. XV
Benzer Tezler
- A challenge to copyright: Text and data mining
Telif hukukunda metin ve veri madenciliği
ESMA MUHEYNE DOĞAN
Yüksek Lisans
İngilizce
2024
Hukukİstanbul Medeniyet ÜniversitesiÖzel Hukuk Ana Bilim Dalı
DOÇ. DR. CAHİT SULUK
- Kişinin görüntüsü üzerindeki hakkının ihlali ve sonuçları
Violation of one's right to the image and its consequences
TUTKU CEREN AKINCI
Yüksek Lisans
Türkçe
2024
HukukAnkara Hacı Bayram Veli ÜniversitesiÖzel Hukuk Ana Bilim Dalı
PROF. DR. ABDULKERİM YILDIRIM
- Halid Ziya Uşaklıgil'in anlatılarında hatırlamanın psikanalitik görüntüsü üzerine bir inceleme
An analysis on the psychoanalytic image of remembering in the narratives of Halid Ziya Uşaklıgil
NACİYE İNCİ
Yüksek Lisans
Türkçe
2024
Türk Dili ve EdebiyatıKarabük ÜniversitesiTürk Dili ve Edebiyatı Ana Bilim Dalı
DOÇ. DR. SELÇUK ATAY
- Kayden şayi mahallen müstakil kullanım durumunda olan alanlarda ıslah imar planı uygulamaları ve sonuçları: Ataşehir örneği
Implementation of development plans in unauthorized subdivision areas and it's results: Atasehir case
SELÇUK AYDEMİR
Yüksek Lisans
Türkçe
2013
Hukukİstanbul Teknik ÜniversitesiŞehir ve Bölge Planlama Ana Bilim Dalı
DOÇ. DR. Ş. ŞENCE TÜRK