Michigan State University. "Facial-recognition technology proves its mettle." ScienceDaily. ScienceDaily, 24 May 2013. <www.sciencedaily.com/releases/2013/05/130524142549.htm>.
It was recently uncovered that Facial-Recognition technology may not be as accurate as we may think. For one, facial recognition software is very picky, if you were to use a picture in different lighting or with a varying facial expression or angle, the software may not pick up on the similarities. The fickle nature of this software means that it has much fewer applications than would would think. According to the article, facial recognition software will only really work if the suspect is facing the camera in the right lighting, otherwise efforts could prove to be unsuccessful. This does not mean that the software is not useful, however, in an experiment at Michigan State University, the software was able to identify one of the Boston Bombers, while the older brother was not able to be identified since he was wearing sun glasses. If it came down to it, the software could be used to identify criminals if a security camera was able to get good enough footage. When the conditions are right, the accuracy of identification is said to be 99%. In addition to facial recognition software, the article also mentions software that could be used to identify and match tattoos, which could prove to be a success if facial recognition is not successful.
I believe that this article is significant to the future of forensics and our society as a whole because the technology described in it can be used to identify criminals and terrorists in the future. If it could be improved, it could prove to be a lot more useful in the future when it really is needed. I found the tattoo identification technology to be useful as well, since it could be used accurately when facial recognition proves to be fruitless.
I found this article very interesting to read, and was impressed by the technology reviewed in it, even if it did have many shortfalls. I found the tattoo identification technology to be very interesting since I didn’t know that it even existed. I think that the topics covered in this article were both relevant and useful to the identification of suspects, and will hopefully prove to make a big difference in the future.
4 comments:
I thought that Aran did a great job of reviewing this article. What stood out to me the most was how thoughtful he was in his writing and how formal and to the point his review was. What also stood out to me was the significance paragraph because he gave key clues as to what this facial recognition profiler can do to our society. Not only that but Aran did a great job of using a real life example of the facial recognition, the Boston bombers.
If I were to give any advice on how the review could have been improved, I think that if Arana addressed this question it could have been beneficial: How could they have gotten the second Boston bomber’s identification if you said that he was wearing sunglasses? My second piece of advice would be to ask: what are some improvements that need to be made to the facial recognition technique?
I very much so enjoyed reading Aran’s review of this article, and what I found interesting was the same thing that Aran found interesting. I thought that it is very cool that the facial recognizer has the potential to identify people by their tattoos.
I thought Aran did an excellent job presenting this article. He discussed using facial recognition software to identify criminals,Facial recognition software seems to be more of a hassle than anything else, even though its 99% accurate under the right circumstances its also almost useless if something as simple as lighting isn't up to standard, leading to the question is it worth it? one thing Aran could do to make this review better is show how costly this technique is, how expensive is it to run these tests and is it worth the money?
Sean Ryan
Forensics
I found Arans article facial recognition proves its mettle to be a great article. I like how it is very relevant to our course. This article reminded me of one I read before which was on fingerprints. Both fingerprints and this face recognition how great potential but can be inaccurate which could put innocent people in jail. I liked how he mentioned how accurate this is when the conditions are right which make me feel that this is something that is very important and can revolutionize police investigation.
I wish he gave us a case on how this facial recognition was negative and also if he could have said how this could be improved i would have felt the summary would have been much better.
I found it interesting how he mentioned tatoo identification which is something that is being improved on and can help catch many criminals.
Michigan State University. "Facial-recognition technology proves its mettle." ScienceDaily. ScienceDaily, 24 May 2013. .
In my opinion, I believe Aran did an amazing job picking a relevant topic that was very interesting as well. One aspect that I liked about this review was how Aran talked about the cons of facial recognition software. He also did a great job explaining WHY it is so important people are aware of the errors that can arise from facial recognition. Lastly, Aran did a great job by making his review have lots of great details which resulted in a great learning experience. Although Aran did do a great job.... there were still places he could've improved on. One of them being how we might be able to fix the errors that can occur in facial recognition software if we can even fix it at all. Another thing I thought Aran should've included was a case where facial recognition resulted in a false verdict and how it had effected his or her life. In all, this was a great article & review to read, with very interesting facts such as how there is also a tattoo verification system too which I thought to be very interesting.
Michigan State University. "Facial-recognition technology proves its mettle." ScienceDaily. ScienceDaily, 24 May 2013. .
Post a Comment