Statistical Analysis to Determine the Feature Resolution of Processed Image Data

uah p 21001pan

Docket: UAH-P-21001


Despite the abundant use of microscopy to characterize material microstructure, standardization of data analysis methods of these image-processing techniques is lacking. Parameters such as scan resolution or spatial resolution are typically reported, but determining the resolvable feature resolution of these scans requires added evaluation. Contemporary methods for evaluating feature resolution fall short due to their dependence on the hardware system and imaging parameters, which requires recalculation for different instrumentation and imaging parameter combinations. 

Researchers at UAH have developed a technology that utilizes data distribution to establish a method of determining the feature resolution of any segmented metric of an acquired dataset. Analysts can use this method to make informed decisions regarding how many images are required for accurate characterization and as a guide to inform instrument selection and setup. 

This technology provides valuable information in determining the capability of an imaging technique to evaluate relevant features and quantifies the associated error of the features. The information obtained by the technology provides a better understanding of the imaging capability compared to the scan resolution. It is advantageous over other methods because it fundamentally accounts for the influence of instrumentation and imaging parameters, and it also allows for estimation of the number of “missing” objects that exist below the resolution of the instrumentation. 


  • Image analysis and data processing
  • Medical imaging (MRI and X-Ray)
  • Materials characterization (Electron microscopy, X-ray computed tomography, optical microscopy)
  • Biology and cell engineering
  • Imaging radar


• Applicable to any dataset that follows a distribution
• Independent of the imaging technique and imaging parameters
• Number of objects below the instrument resolution may be inferred


Rotational Swing System and Methods for Providing Vestibular Stimulation