Oldal 111 [111]
OCR
3. Image post-processing The importance of developing post-processing methodologies to aid in the interpretation of luminescence and other multispectral images is increasingly being recognised, both as a means to extract the maximum information from the images but also to standardise images and make them device-independent, allowing them to be compared more directly. Many of the approaches reported’? have included corrections for some of the experimental factors and phenomena discussed in Chapter 1. Some have included existing nip2 tools (see below), such as that used to carry out colour corrections based on use of a Macbeth target. Others have proposed similar workflow approaches but have built these into proprietary software programs, such as Matlab, which require the purchase of these specialist programs and considerable knowledge of their use by the user. In almost all cases, the postprocessing methodologies have been slow to implement, are in many cases at least partly Subjective and require the use of several different software programs making them unsuitable for wide adoption by a broad range of users with varying understanding of the mathematical processes involved. The current approach has developed a unified set of user-friendly tools housed in a dedicated ‘workspace’ for nip2, the graphical interface of the free image processing system VIPS° (see below). This chapter describes how the transformations outlined described in Chapter 1, and the data collected as a result of the acquisition protocols outlined in Chapter 2, have been combined in a nip2 workspace to form a series of image correction workflows. Prior to carrying out any post-processing of the images recorded however, these must first be correctly transferred from the camera and converted to TIFF files from their native digital camera RAW formats. Instructions on how to carry this out are provided below, together with instructions on how to download and install the nip2 software and the workspace developed to undertake the post-processing. Finally, the four workflows for the correction of the multispectral image sets as proposed in Chapter 1 and collected as described in Chapter 2, are described and a full set of instructions for their use is provided, together with the postprocessing requirements (the images or other information) required to complete them. The workflows are considered in terms of the two categories of images established in previous chapters: reflected and photo-induced luminescence images. Some quick reference instructions, intended for use once familiarity with the operation of the workspace and the various workflows for the correction of the image types described has been achieved, are included in Appendix 3. a. Transferring images from the camera and converting from RAW As discussed briefly in Chapter 1, whenever a captured image is converted into a standard JPEG or TIFF file by the camera or the RAW image conversion software (such as Adobe Photoshop or the software from the camera manufacturer), a number of processes are carried out to the data from the sensor. These are summarised in Figure 3-1. Version No. 1.0 103 Date : 14/10/2013
