Abstract
A dominant theme in vision research is that important characteristics of the visual pathway evolved to be effective in processing natural scenes. Given this perspective, one can learn about the nature of visual processing from a quantitative analysis of natural images. Such analysis can benefit from the camera as a measuring device. As such, the camera should not introduce arbitrary artifacts into the image formation process. This paper describes how to correct a number of unnecessary artifacts associated with obtaining natural scene statistics with a digital camera. For artifacts that are inherently part of image formation, and where elimination is not possible or appropriate, we describe methods for transformation and quantification.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 30-42 |
| Number of pages | 13 |
| Journal | Journal of the Optical Society of America A: Optics and Image Science, and Vision |
| Volume | 26 |
| Issue number | 1 |
| DOIs | |
| State | Published - Jan 1 2009 |
Fingerprint
Dive into the research topics of 'Camera calibration for natural image studies and vision research'. Together they form a unique fingerprint.Cite this
- APA
- Standard
- Harvard
- Vancouver
- Author
- BIBTEX
- RIS