Quantitative Analysis of Real-Time Image Mosaicing Algorithms
| Autor: | Nemra Abdelkrim, Sayah Houari, Khellal Atmane |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Computer science
business.industry Distortion (optics) Image registration 020207 software engineering Field of view 02 engineering and technology Image (mathematics) Aerial photography Quantitative analysis (finance) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Computer vision Artificial intelligence business Homography (computer vision) |
| Zdroj: | IWSSIP |
| DOI: | 10.1109/iwssip.2018.8439448 |
| Popis: | Image registration and photo-mosaicing of related imagery from unmanned aerial vehicles (UAVs) is an active research topic. Aerial photography and mosaicing became crucial for surveillance and reconnaissance. It consists in aligning multiple images to construct a single large image of a 3D scene allowing the operator to view images that offers a wider field of view than standard images. Offline registration and mosaicing of collected images from UAVs has proven to give great results but is numerically intensive and somewhat slow. In this paper, a new approach for real time mosaicing is proposed based on new registration approach which reduces accumulated error and distortion of the current image. A quantitative analysis of the new approach is performed. This analysis allows the comparison of two images mosaic using some parameters to evaluate the quality, the Distortion rate and the speed of mosaicing. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|