Feature extraction: percentage of missing values by row

A good quality of 3d scans is crucial for an assessment of similarity of striation marks between different bullets. Here, we measure the percentage of missing values in the scan. Generally, a higher percentage of missing values is associated with a lower scan quality.

extract_na_row(x3p)

Arguments

x3p

scan in x3p format

Value

percentage of missing values in each row of the scan's surface matrix

Examples

data(fau277_bb_l2) # bad scan
nas_bad <- extract_na_row(fau277_bb_l2)
plot(nas_bad)

# interesting pattern: we would expect a similar pattern for all scans
# (going from very low percentage of missing values at the top of a scan to a
# fairly high percentage along the bottom)

nas_good <- extract_na_row(fau001_ba_l1) # excellent scan
plot(nas_good) # similar pattern to above, but low values for most of the scan