Automated SNR Search to Support an Audit

Abstract

In this paper, we describe how we used MATLAB(registered trademark) to measure the signal to noise ratio in a stack of radiographs of M1 shells. We created a MATLAB program that converted each image into a matrix of gray values corresponding to each pixel. Using the techniques of a previous technical report, we identified three regions of interest dictated by military specifications. We then recorded statistics associated with each region. The program successfully recorded the average and maximum SNR of each of the three regions of interest. Our methodology allows for more audits to be performed of our production facilities. It also allows for more objectivity when assessing image quality. Finally, the techniques described here allow for potential critical escapes to be quickly identified and brought to the attention of Level 3 radiographers and program managers.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2020
Accession Number
AD1094253

Entities

People

  • Ruben A. Favaro
  • Walter S. Rose

Organizations

  • Picatinny Arsenal

Tags

Communities of Interest

  • Human Systems
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence Software
  • Computer Vision
  • Computers
  • Convolutional Neural Networks
  • Detection
  • Horizontal Orientation
  • Information Science
  • Machine Learning
  • Measurement
  • Neural Networks
  • Orientation (Direction)
  • Recognition
  • Standardization
  • Standards
  • Statistics
  • Vertical Orientation

Readers

  • Image Processing and Computer Vision.
  • Software Engineering
  • Systems Analysis and Design