prikaz prve stranice dokumenta Interpolacija slike uz očuvanje rubova i implementacija rješenja na realnu ADAS platformu
  Download
PDF 4.93 MB
master's thesis
Interpolacija slike uz očuvanje rubova i implementacija rješenja na realnu ADAS platformu
2021. urn:nbn:hr:200:220197
Kelava, Božidar
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Communications
Chair of Multimedia Systems and Digital Television

Cite this document

Kelava, B. (2021). Interpolacija slike uz očuvanje rubova i implementacija rješenja na realnu ADAS platformu (Master's thesis). Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:220197

Kelava, Božidar. "Interpolacija slike uz očuvanje rubova i implementacija rješenja na realnu ADAS platformu." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2021. https://urn.nsk.hr/urn:nbn:hr:200:220197

Kelava, Božidar. "Interpolacija slike uz očuvanje rubova i implementacija rješenja na realnu ADAS platformu." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2021. https://urn.nsk.hr/urn:nbn:hr:200:220197

Kelava, B. (2021). 'Interpolacija slike uz očuvanje rubova i implementacija rješenja na realnu ADAS platformu', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 20 October 2021, https://urn.nsk.hr/urn:nbn:hr:200:220197

Kelava B. Interpolacija slike uz očuvanje rubova i implementacija rješenja na realnu ADAS platformu [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2021 [cited 2021 October 20] Available at: https://urn.nsk.hr/urn:nbn:hr:200:220197

B. Kelava, "Interpolacija slike uz očuvanje rubova i implementacija rješenja na realnu ADAS platformu", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:200:220197

Islandora Bookmark - Please login to use this feature
Metadata
TitleInterpolacija slike uz očuvanje rubova i implementacija rješenja na realnu ADAS platformu
Title (english)Image interpolation with edges preserving and solution implementation on a real ADAS platform
AuthorBožidar Kelava
MentorMario Vranješ (mentor)
Committee memberMario Vranješ (predsjednik povjerenstva)
Committee memberRatko Grbić (član povjerenstva)
Committee memberMarijan Herceg (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Communications)
(Chair of Multimedia Systems and Digital Television)
Osijek
Defense date and country2021-09-16, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Telecommunications and Informatics
Abstract
U ovom radu opisane su postojeće metode neprilagodljivih algoritama za interpolaciju slike. Objašnjen je problem interpolacije slike, opisani su matematički modeli triju metoda za interpolaciju slike (metode najbližih susjeda, bilinearne interpolacije i bikubične interpolacije) te je dan pregled te je dan pregled postojećih rješenja. Opisana je potom vlastita implementacija triju navedenih metoda interpolacije u C programskom jeziku na računalu. Nakon implementacije u C programskom jeziku na PC-u odabrane metode su implementirane na ADAS Alpha ploči, također koristeći C programski jezik. Nakon same implementacije na ploči, pažnja je posvećena raspodjeljivanju posla na dostupne procesora na Alpha ploči. Nakon optimalne raspodjele posla na pojedine procesore, uspoređivane su performase interpolacije trima spomenutim metodama prilikom izvođenja na ugradbenoj ADAS Alpha ploči. Rezultati su pokazali da na odabranom skupu od 50 testnih slika, od kojih je 30 automotiv sadržaja, 10 sadržaja motiva iz prirode s više i manje detalja i 10 umjetno generiranih slika, najbolji kompromis, ukoliko se u obzir uzmu vrijeme izvođenja i kvaliteta interpolirane slike, daje bilinearna interpolacija i to kada se zadatak interpolacije raspodijeli na tri različita procesora (DSP1, DSP2 i A15) prilikom povećavanje rezolucije slike četiri puta, odnosno s 640×360 elemenata slike na 1280×720 elemenata slike, gdje je prosječno vrijeme izvođenja 280,16 ms. Ispitivanjima je utvrđeno da za smanjivanje rezolucije slike četiri puta, odnosno s 640×360 elemenata slike na 320×180 elemenata slike, procesor A15 ima najkraće vrijeme izvođenja od 53,4 ms.
Abstract (english)
This paper describes the existing methods of non - adaptive interpolation image algorithms. The problem of interpolation images is explained, mathematical models of three methods for interpolation images (nearest neighbor methods, bilinear interpolation and bicubic interpolation) are described, and an overview is given and an overview of existing solutions is given. Then, the own implementation of the three mentioned interpolation methods in the C programming language on a computer is described. After implementation in the C programming language on a PC, the selected methods were implemented on the ADAS Alpha board, also using the C programming language. Following the same on-board implementations, attention was paid to allocating work to the available processors on the Alpha board. After optimal distributions, individual procedures were sent, and the interpolation performance of the three mentioned methods was compared when performed on a built-in ADAS Alpha board. The results showed that in the selected set of 50 test images, of which 30 automotive content, 10 nature motifs with more and less detail and 10 artificially generated images, the best compromise, if we take into account the execution time and quality are interpolated image, gives bilinear interpolation when the interpolation task is divided into three different processors (DSP1, DSP2 and A15) when increasing the image resolution four times, ie from 640 × 360 image elements to 1280 × 720 image elements, where the average execution time is 280 , 16 ms Tests have found that to reduce the image resolution four times, from 640 × 360 elements to 320 × 180 element images, the A15 processor has the shortest execution time of 53.4 ms.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:220197
Study programmeTitle: Graduate University Study Programme in Electrical Engineering; specializations in: Communications and Informatics, Power Engineering
Course: Communications and Informatics
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike (magistar/magistra inženjer/inženjerka elektrotehnike)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2021-10-04 14:07:50