{"title":"Fast mosaicing method based on image resizing pre-processing","authors":"A. Laraqui, Mohammed Laraqui, A. Saaidi","doi":"10.24867/jged-2023-1-037","DOIUrl":null,"url":null,"abstract":"In recent years, mosaic images have found great success thanks to the increasing development in the field of imaging as well as the technological evolution of computer systems (camera, mobile, etc.). Mosaic images are obtained by merging several images of the same scene. The process incorporates several steps, each of which requires resources and execution time depending on the size, quality and resolution of the images used. In this paper, we propose a new image mosaic method that significantly reduces the execution time. The idea is to apply the stages of registration and search for the best inliers, necessary for the calculation of the geometric transformation, to the miniature of images. This allows the minimization of the overall processing time without altering the quality of the results. The experiment, on a database of images, shows that the proposed algorithm provides rapid results compared to similar methods. Also, we have extended our method to generate 360° panoramic images.","PeriodicalId":16019,"journal":{"name":"Journal of graphic engineering and design","volume":" ","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of graphic engineering and design","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.24867/jged-2023-1-037","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Engineering","Score":null,"Total":0}
引用次数: 0
Abstract
In recent years, mosaic images have found great success thanks to the increasing development in the field of imaging as well as the technological evolution of computer systems (camera, mobile, etc.). Mosaic images are obtained by merging several images of the same scene. The process incorporates several steps, each of which requires resources and execution time depending on the size, quality and resolution of the images used. In this paper, we propose a new image mosaic method that significantly reduces the execution time. The idea is to apply the stages of registration and search for the best inliers, necessary for the calculation of the geometric transformation, to the miniature of images. This allows the minimization of the overall processing time without altering the quality of the results. The experiment, on a database of images, shows that the proposed algorithm provides rapid results compared to similar methods. Also, we have extended our method to generate 360° panoramic images.