Why you should not use image files for data analysis
Image logging and export formats in BeamGage are intended for use as easily created screenshots for reports and presentations. Alternatively there are real data formats for use in mathematical analysis, (e.g. HDF, ASCII csv). Matlab does have native support for HDF formatted data. If you saved and bgData files then you will have full fidelity HDF data files.
Here are two significant problems of attempting data recovery on image files:
First is data loss in the form of the palette. All of the palettes in BeamGage are custom designed palettes that provide optimal visibility of the dynamic range for various conditions. None of them are linear intensity palettes. The highest intensity resolution you can get is approximately 7-bits (there are a maximum of 128 colors, additionally we add a zero value color and a negative value color). The 128 colors are mapped across the dynamic range of the data. 4096 counts for 14-bit data, 255 counts for 8-bit data, and so on. Each bracket is then assigned a color value. Therfore, depending on the original format you may have severe data loss in your situation.
Secondly, the 2D beam display cannot, under most frame formats, be mapped directly to monitor display pixels. The typical default frame format of 1600x1200 would consume most of your average monitor display. Due to this the screen captures you obtained will not have retained the complete pixel detail. The loss will depend on the size of the display at the time of export. You will also have data loss due to any overlays such as the apertures, cursor and crosshair, and 1D profiles.