Overview.md

+[[_TOC_]]
+
+## About Gepard
+
+
+Gepard is a software specifically developed for particle-based Raman/FTIR analysis of microplastics. However, it can be used for any application that requires detection and characterisation of multiple particles on a substrate. 
+It includes two modules that handle the measurement (gepard, and/or the control software of the instrument) and the data evaluation (gepardevaluation) part of the analysis separately.
+
+
+## Supported instruments
+
+Currently, GEPARD supports working with following instruments on varying levels of interaction:  
+- Confocal Raman microscopes:
+ - alpha300 R by WITec 
+ - QONTOR by Renishaw 
+- FTIR microscopes:
+ - Spotlight 400 by PerkinElmer
+ - Nicolet by ThermoFisher
+- Optical microscopes:
+ - Axio Imager.Z2m by Zeiss
+
+
+Feel free to expand Gepard for use with other instruments. Prerequisite is an instument with motorized xyz-stage and control software that allows external client commands.
+
+
+## Gepard measurement in a nutshell
+
+In short, the analysis can be broken down to following steps:  
+
+- Measurement  
+  0. Your sample preparationshould result to particles on a planar substrate. Silicon wafer filters are the best solution, because they are perfectly planar and shiny, so they appear black in darkfield microscopy, and they are well-suited as substrates for Raman and FTIR measurements. On their downside, they are expensive and fragile.
+  1. Image aquisition. The region of interest is scaned at various focus levels (z-stack). The images are used to create a height map and a maximal focus image of the sample.
+  1. Partice detection. Using conventional image processing tools or a neural network algorithm the particles are detected on the image and their positions are calculated.
+  1. Spectrum scan. The particle coordinates are passed to the microscope along with measurement paramenters for automatic measurement.
+- Evaluation  
+  1. Import spectra.
+  1. Spectra evaluation or results import. Evaluation of the spectra can either be done in gepardevaluation or externally with a tool of your choice. If spectra evaluation was performed with an external tool, the results need to be imported in gepardevaluation from an ASCII file.
+  1. Review of the results. This feature is what makes Gepard stand out. After the dataset has been completed, the user has complete control of the automatically produced results. Particle contours and properties, and spectra assignements can be corrected manually as needed.
+  1. Export of the results. Particle results and sample statistics can be exported as xlsx files. In addition, gepardevaluation provides an interface for uploading results to SQL databases.
+
+
+## Gepard project structure
+
+Each project is stored in its own folder, which is automatically created when creating a new project. The folder contains the following:
+- pkl file. This is the core of the project, where all book-keeping is done. It contains information about the project state (e.g. which stages have been completed), the sample (e.g. coordinate system), the particles (e.g. coordinates, contours, etc.), completed measurements, assigned results, and all relations between them.
+- pkl backup files. These are created by gepard and gepardevaluation after a number of actions have been performed on the dataset.
+- zvalues.tif is the height map of the sample in the form of a grayscale image.
+- spectra folder. Loaded spectra series are stored here as numpy arrays. (one .npy file per series)
+- tiles folder. It contains the optical image of the sample in segments at various zoom levels. They are used to dynamically load the image in gepard or gepardevaluation.
+- scanimages folder. This is used for temporary storage of the optical images during image acquisition. It should be empty, unless the software crashed during acquisition.
+- tmp.bmp is an auxiliary file created during the image acquisition.
+
+
+
+
+