INVESTIGATION OF IMAGE MAGNIFICATION METHODS
Journal Title: Jaunųjų mokslininkų darbai - Year 2014, Vol 42, Issue 2
Abstract
Image magnification is a process by which the image is magnified only visually. Image magnification is applied in many different areas. The images are magnified to make a diagnosis in medicine, to identify people or objects in observation systems or police work, etc. Images are magnified by using different methods. These methods can be categorized as adaptive or non-adaptive. Non-adaptive methods magnify all pixels by using the same method while adaptive methods choose the best method for the part of the image. The aim of the research is to experimentally compare image magnification methods while enhancing synthetic and real images. The objectives of the research are to create a methodology for research and to compare image magnification methods by using selected methods (nearest neighbour interpolation, bilinear interpolation, bicubic interpolation, Lanczos-2 and Lanczos-3). Methods are evaluated by calculating the mean square error, mean absolute error, correlation coefficient, peak signal-to-noise ratio and the average of untrue pixels. Methods of image magnification are also aligned by performing a two-sample Student’s t-test. This investigation shows that magnification methods can be aligned independent of reduction methods. Best results are seen when the image is reduced using Lanczos-3 method. It is also deduced that the choice of best magnification method does not depend on how many times the image was magnified or what the type of image is, as the results are identical for synthetic and real images. Key words: image magnification, synthetic and real images, evaluation of magnification methods, Lanczos-3.
Authors and Affiliations
Kristina Vyšniauskaitė, Nerijus Ramanauskas
Keywords
Related Articles
- EP ID EP159372
- DOI -
- Views 108
- Downloads 0
GLOBALIOS OPTIMIZACIJOS ALGORITMAS, NAUDOJANTIS LOKALŲ LIPŠICO KONSTANTOS ĮVERTĮ
Šiame darbe pasiūlytas naujas deterministinis globalios optimizacijos algoritmas, skirtas juodos dėžės funkcijoms, kurioms galioja Lipšico sąlyga, bet Lipšico konstanta nežinoma. Algoritmo pradinėje stadijoje leistinoji...
DUOMENŲ IŠRINKIMO INTERNETO PUSLAPIUOSE ALGORITMAS, PAREMTAS DUOMENŲ TARPUSAVIO PANAŠUMU
Straipsnyje aprašoma ir analizuojama duomenų išrinkimo iš interneto puslapių problema ir egzistuojantys sprendimai. Atsižvelgiant į šiuo metu vyraujančias tendencijas duomenims išrinkti naudoti ne keliu, o duomenimis par...
MOTERŲ, AUGINANČIŲ VAIKUS, SUNKUMAI DERINANT ŠEIMĄ IR KARJERĄ
Šeimos ir karjeros derinimas – kiekvienam žmogui aktuali problema, tačiau dažniausiai su tuo susiduria moterys, auginančios vaikus. Visuomenėje tebevyrauja stereotipas, kad moteris turi aukotis dėl šeimos. Išanalizavus m...
Minimally Invasive Surgery Service Development System in Šiauliai Republican Hospital
The paper aims to discus the theoretical aspects of health care reform and further development features, provides the concept of health care services, describes the concept of minimally invasive surgery and the developme...
ANALYSIS OF CUSTOMERS‘ ACTIONS IN LITTLE-VISITED E-SHOPS
The purpose of this study is to investigate the use of a statistical analysis of e-shop traffic and to demonstrate how to integrate a fairly simple web data collection and analysis system into an already functionin...