Digital image data is increasingly playing an important role in metrology. This project aims to produce better measurements from (2D) image data, measurements that are more reliable, robust and traceable by considering meta-data (data about how the image is made). The work benefits many areas of NMS and NPL science that use measurements from image data.

• MONAT (Measurement of Naturalness) project, which is co-funding WP4 of project 3.1, to investigate the image-based measurement of other natural (or purported natural) subject
  
  
• Work on freeform Coordinate Measuring Machines (CMM)
  
  
• Stereoscopic imaging to measures distances such as cloud height measurement
  
  
• Biometric image quality: extending the work examining iris image quality undertaken under the MET programme

Investigation of current methods for deriving measurement data from 2D images to determine where the measurement process fails in repeatability, robustness and/or traceability; and to determine sources of uncertainty in the measured quantities. This will build on core SSfM activities of modelling, uncertainty evaluation and signal processing. The output of biometric systems is often an accept/reject decision, rather than a numerical measurement, so there are also techniques from project 3.1 on "Inference and decision making", and previous work on data fusion, that will apply. These SSfM techniques will then be used to improve the performance of measurement processes from image-based data. This work will take examples from the NPL & NMS projects and provide generic techniques that may be applicable in those projects.