Classification Workflow
Adding Reference Spectra
Add References from the Results
10 min
background amics offers three ways to add refrence spectrum from data that has already been collected seed search that will use a single selected spectrum as a "seed" to search the entire measurement seed search is inituated by either hovering over a measurement and right clicking or clicking a single x ray measurement point and right clicking and selecting "generate material refrence spectra" select one or more spectrum from the measuremnt and righ click to select "generate material refrence spectra from selection" this will only add the sum spectra and will not seed search use the docid\ kbj1vfvvvp1hmrkbkk92o to generate a bin and add the contents of the bin as a sum sprectrum mateiral refrence spectrum adding refrence spectrum from the image in addition to the acquisition of reference spectra directly from the sem eds system, amics also allows for the spectra collected during analysis to be used at the reference spectrum to access the spectra data within process use the "xray points" option in the main menu ribbon, and the main resultant panel should be in "images" and needs to be zoomed in close enough for the x ray cross hair to appear once the x ray crosshairs are visible, it is possible to visualize the spectra by rolling over the crosshair this will bring up a subwindow at the base of the main interface showing the spectrum useing seed search seed search with search the entire measurement to generate a refrence spectrum that has enough counts and captures the background correctly once the appropriate spectrum has been selected, right clicking on the spot will pop up a menu with the option to "generate material definition" this process can then be repeated to add additional reference spectra to refine the final material classification method amics uses a unique seed searching algorithm to search for similar spectra to the selected spectrum in the same dataset and accumulate them into a summed spectrum to be used as the reference spectrum the advantage of using a summed spectrum is to accommodate slight variations between points, as well as improve the signal to noise ratio of the final reference spectrum the default setting for amics is to accumlate similar spectra until a summed spectra containing at least 10,000 total x ray counts is achieved however, this target value can be adjusted in the settings options in the process (seed search target count) amics will then accumulate to close to this target total count, and it is required that this value is larger than the standard counts (low) setting use selection selecting one or more segments can also be used to generate a refrence spectrum select more than one spectra that seem the same this can be done by holding the shift key and dragging rememeber, if the segment is select it will select a single x ray grain and particle layers will select all x rayes in that layer from this the user can select "generate mineral refrence spectrum from selected" and a sum spectra will be generated note shift + drag will select one or more items shift + esc to unselect everything options in the contect menu to unselect a single item making an x ray "active" by clicking it gives the option for either use seed search or to only use the selection be cautious! selecting can be a great way to selectivaly add material refrence spectra however, as the video shows, this is not always a great idea a single "outlier" selection from a small segment can cause extra peaks to show up in a artifical spectrum oops adding refrence spectrum from spectrum tree amics spectrum tree is a unique tool that can be used to filter and cluster spectra based on user defined requirements the tool is unique to amics as the amics acquisition process collects and retains the spectra from every x ray analysis spot however, an additional feature of spectrum tree is to allow users to generate reference spectra based on clustering results to generate reference spectra from a resultant clustering or filter from spectrum tree, right clicking the resulting cluster will bring up a pop up menu on this pop up menu select "make new standard", amics will then use the average spectra (thick dark blue line) to create the reference spectra, which are saved to the selected classification method
