Loading...
Please wait, while we are loading the content...
Similar Documents
Image analysis for microscopy screens Image analysis and processing with EBImage by
| Content Provider | Semantic Scholar |
|---|---|
| Author | Sklyar, Oleg |
| Copyright Year | 2006 |
| Abstract | The package EBImage provides functionality to perform image processing and image analysis on large sets of images in a programmatic fashion using the R language. We use the term image analysis to describe the extraction of numeric features (image descriptors) from images and image collections. Image descriptors can then be used for statistical analysis, such as classification, clustering and hypothesis testing, using the resources of R and its contributed packages. Image analysis is not an easy task, and the definition of image descriptors depends on the problem. Analysis algorithms need to be adapted correspondingly. We find it desirable to develop and optimize such algorithms in conjunction with the subsequent statistical analysis, rather than as separate tasks. This is one of our motivations for writing the R-package EBImage. We use the term image processing for operations that turn images into images, with the goals of enhancing, manipulating, sharpening, denoising or similar (Russ , 2002). While some image processing is often needed as a preliminary step for image analysis, image processing is not the primary aim of the package. We focus on methods that do not require interactive user input, such as selecting image regions with a pointer etc. Whereas interactive methods can be extremely effective for small sets of images, they tend to have limited throughput and reproducibility. EBImage uses Magick++ interface to the ImageMagick (2006) image processing library to implement much of its functionality in image processing and input/output operations. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://bioconductor.org/packages/1.9/bioc/vignettes/EBImage/inst/doc/AnalysisWithEBImage.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |