Demo of target-analyzer, photo-based scoring of paper targets, from this case study. Synthetic targets and readings. On the held-out target, cv_blob and yolo at 0.10 and 0.25 are matched to the real measured run; every other number is illustrative.

target-analyzer

demo range

2026-08-25 · shooter A · 10 m

readingproposedmatchedmissedspuriousoffset
cv_blob86421.02 mm
yolo ≥ 0.2599100.51 mm

move the threshold

yolo cut atproposedmatchedmissedspuriousoffset
yolo ≥ 0.051010000.51 mm
yolo ≥ 0.10 (the default)1010000.51 mm
yolo ≥ 0.151010000.51 mm
yolo ≥ 0.2599100.51 mm
yolo ≥ 0.4088200.56 mm

A detector only proposes: a person confirms every target, so a spurious hole costs one click to remove and a missed one costs a click to place. Lowering the threshold trades misses for spurious holes. 0.10 is the default the real app ships with. Two readings are the same hole within 4 mm.

confirmed by a person

ring 10ring 9ring 10ring 9ring 8ring 8ring 8ring 8ring 8ring 9

the truth every reading is scored against

cv_blob

ring 9ring 9ring 9ring 8ring 8ring 9ring 2ring 2

classic blob detection, no threshold to move

yolo ≥ 0.25

ring 10ring 9ring 10ring 9ring 8ring 8ring 8ring 8ring 9

the trained detector, cut at the threshold chosen above