Correction: Video Analysis Math

Correction: Reexamining the math
in a recent forensic video analysis article

The July-August 2009 issue of Evidence Technology Magazine included an article, “Matching Video to its Source” (Volume 7, Number 4, pages 10-13). Article contributor Scott Kuntz is a deputy sheriff with the Dane County Sheriff’s Office in Madison, Wisconsin. In the article, Kuntz explained the process of matching evidence videotape to the camcorder that originally recorded it. The technique involved locating and mapping bright spots in the evidence footage. These bright spots were caused by failed CCD sensors in the camcorder that originally recorded that footage. Camcorders from the suspect’s residence were subsequently examined in an effort to document failed CCD sensors in the same fashion. After a match had been found, the probability of finding another sensor in that same class, with the same exact defects, was calculated. Kuntz calculated the probability of finding another sensor array in that same class with the same defects would be 1 in 8,100,000,000,000,000,000,000.

In September 2009, a reader of the article (Robert Blackledge, a forensic chemist consultant) contacted ETM editor Kristi Mayo. The reader pointed out that the method used in the article to calculate the probability was wrong. Mayo in turn contacted Kuntz.

Kuntz consulted with a mathematics professor and a statistician about the calculation. They confirmed the opinion of the concerned reader. They agreed that the calculation should have been done as follows. If a sensor array had a total of 300,000 sensors and 4 of them had failed, calculating the probability of finding another like it could be achieved with the following formula:

300,000 x 299,999 x 299,998 x 299,997
4 x 3 x 2 x 1

The corrected probability of finding another sensor array in that same class with the same defects is:

1 in 337,493,250,041,250,000,000

It is important to note that Kuntz did correctly compare and identify the failed sensors in the evidence footage and test recordings from the seized cameras. However, Kuntz acknowledges his error in calculating the probability of finding another camera in the same class with the same defects. Kuntz did have his original work double peer-reviewed by qualified forensic video analysts. He learned that he should have had his mathematics peer reviewed by probability and statistics professionals, as well.

Kuntz regrets the possibility that his mathematics error might have misled any reader of this magazine. This correction has been made so others might learn as well.

Deputy Sheriff Scott Kuntz can be reached by e-mail at: This e-mail address is being protected from spam bots, you need JavaScript enabled to view it

Have any thoughts to contribute?
Send an e-mail to the editor:

This e-mail address is being protected from spam bots, you need JavaScript enabled to view it


ORIGINALLY PUBLISHED:
"News from the Field"
September-October 2009 (Volume 7, Number 5)
Evidence Technology Magazine
Buy Back Issue

 
< Prev   Next >






Digital-Image Management at Mass Gravesites

SKELETONIZED REMAINS that were carefully unearthed from the desert sands of Iraq tell their own story: the bones of an adult, still dressed in a woman’s apparel, lie supine. The skull is perforated by a bullet hole. Tucked in the space between the ribs and the left humerus is a much smaller skeleton, bones in the skull un-fused, and the fully clothed body partially swaddled in a blanket.

Read more...