Automatinis transporto priemonių registracijos numerio atpažinimas atlieka svarbų vaidmenį daugelyje programinių paketų, taikomų tiek valstybinėse institucijose, tiek ir privačiose kompanijose, kuriuose yra naudojamos įvairios atpažinimo algoritmų technologijos. Tačiau net ir dabar neįmanoma sukurti idealiai veikiančios sistemos, kuri palieka laisvę efektyviausių algoritmų paieškai. Šio tiriamojo darbo tikslas yra išanalizuoti alternatyvius automobilio numerio lokalizacijos ir kitų atpažinimo etapų algoritmus, jų efektyvumą bei adaptyvumą. Analizė atliekama juos realizuojant ir atliekant tyrimus su testiniais duomenimis bei iš jų gautais rezultatais. Iš realizuotos alternatyvių atpažinimo algoritmų sistemos gauti rezultatai parodė, kad kiekviena atpažinimo proceso grandis yra jautri įvairiems faktoriams, kurių kitimas lemia tarpinių bei galutinių rezultatų variaciją. / Automatic license plate recognition plays an important role in numerous applications and a number of techniques have been proposed for public institutions or private companies. However, even now it is impossible to design a perfect and operational recognition system. It still leaves a space for creativity and research of the most effective algorithms. The main objective of this dissertation is to analyze alternatives of licese plate localization and other stages of recognition, their efficiency and adaptability. Selected means of this research are such as implementation of algorithms, analysis of testing data and test results. Every stage of recognition process is extremely sensitive to different factors which determinate variation of transitional and final results. This was proven by analysis of alternative algorithms functionality.

Automatinis transporto priemonių registracijos numerio atpažinimas atlieka svarbų vaidmenį daugelyje programinių paketų, taikomų tiek valstybinėse institucijose, tiek ir privačiose kompanijose, kuriuose yra naudojamos įvairios atpažinimo algoritmų technologijos. Tačiau net ir dabar neįmanoma sukurti idealiai veikiančios sistemos, kuri palieka laisvę efektyviausių algoritmų paieškai. Šio tiriamojo darbo tikslas yra išanalizuoti alternatyvius automobilio numerio lokalizacijos ir kitų atpažinimo etapų algoritmus, jų efektyvumą bei adaptyvumą. Analizė atliekama juos realizuojant ir atliekant tyrimus su testiniais duomenimis bei iš jų gautais rezultatais. Iš realizuotos alternatyvių atpažinimo algoritmų sistemos gauti rezultatai parodė, kad kiekviena atpažinimo proceso grandis yra jautri įvairiems faktoriams, kurių kitimas lemia tarpinių bei galutinių rezultatų variaciją. / Automatic license plate recognition plays an important role in numerous applications and a number of techniques have been proposed for public institutions or private companies. However, even now it is impossible to design a perfect and operational recognition system. It still leaves a space for creativity and research of the most effective algorithms. The main objective of this dissertation is to analyze alternatives of licese plate localization and other stages of recognition, their efficiency and adaptability. Selected means of this research are such as implementation of algorithms, analysis of testing data and test results. Every stage of recognition process is extremely sensitive to different factors witch determinate variation of transitional and final results. This was proven by analysis of alternative algorithms functionality.

Thesis (Ph. D.)--University of Texas at Austin, 2006. / Vita. Includes bibliographical references.

Univ., Diss.--Frankfurt (Main), 2005. / Literaturverz. S. IX - XXV.

Thesis (doctoral)--Universität Bern, 2009 / Includes bibliographical references (p. xxvii-xxxviii).

Thesis (Ph. D.)--Harvard University, 1993. / Includes bibliographical references (leaves 105-111).

Thesis (M.A.)--University of North Carolina at Chapel Hill, 2004. / Includes bibliographical references (leaf 102). Also available online (PDF file) by a subscription to the set or by purchasing the individual file.

Thesis (doctoral)--Université de Lausanne. / Includes bibliographical references (p. [347]-378) and index.

Thesis (M.A.)--University of North Carolina at Chapel Hill, 2004. / Includes bibliographical references (leaf 102).

Copyright licences can be implied when the doing of a restricted act is covered neither by the express licence of the copyright owner, nor by one of the statutory limitations and exceptions. The manner in which copyright licences are implied, therefore, holds the key to broadening the scope of permissible acts. In contrast to the rigidity of statutory limitations and exceptions, implied licences are more malleable in being able to respond to a diverse set of circumstances, as the need arises. Thus, implied licences can serve as a flexible and targeted mechanism to balance competing interests, including those of copyright owners and content users, especially in today’s dynamic technological environment. However, implication as a process is contentious, and there are no established rules for implying a licence. Implication of a copyright licence is even more complicated because the conceptualisation of a copyright licence is unsatisfactory. The resulting uncertainty has prevented implied licences from being embraced more readily by the courts. The objective of this thesis is, therefore, firstly, to reconceptualise a copyright licence that is broad enough to accommodate the diverse circumstances in which copyright licences arise, and certain enough to assist in finding their constituents; secondly, to propose frameworks for implying copyright licences in a methodical and transparent manner, based on three sources: the consent of the copyright owner; an established custom; and state intervention to achieve public policy goals. The frameworks are also customised differently for implied bare and implied contractual licences. The thesis demonstrates the robustness of these frameworks by rationalising them with the existing case law. Underscoring the contemporary relevance of implied licences, in conclusion, the thesis tests and validates the frameworks in relation to three essential and ubiquitous functions on the internet – browsing, hyperlinking and indexing.