So, we can identify a tiger by his or her pattern. But how do we get to see a tiger? They are naturally very secretive unless forced to approach us through lack of habitat caused by us (see for example cheetah habitat). Well, as we know, there are things called camera traps. These are cameras that are set up and left to be fired automatically by the tiger when it passes in front of the camera. Great, but the photographs will be from different angles and so on. This means a careful study, I suspect of photographs of known animals and a careful comparison made of the newly photographed tiger. A time consuming business, no doubt. However, bearing in mind these factors:
- the importance of conserving the tiger;
- the great importance in reducing trafficking in tiger parts including skins;
- the increased use of camera traps and therefore a larger database of images,
A smart bunch of scientists have come up with a computer program based on two algorithms (complex formula - "algorithm is a finite sequence of instructions, an explicit, step-by-step procedure for solving a problem, often used for calculation and data processing" - src: Wikipedia) that can compare images scanned in to a computer with images already on the computer in the form of a database to make an assessment as to similarity and to take into account all the variables as to image quality and position etc. a barrier to accurate identification.
As I understand it the system has a high level of reliability in matching similar patterns under what is variable conditions and angles. The creators of the system:
- Lex Hiby
- Phil Love
- Narendra Patil,
- N. Samba Kumar
- Arjun M Gopalaswamy
- K. Ullas Karanth,
I would be very interested to know how or if this computer program has been used in apprehending the traffickers in tiger parts. Thanks to Biology Letters.
Tiger Patterns and Tiger Stripes to Bengal Tiger Facts
Tiger Patterns and Tiger Stripes -- Photo is by SubyRex and published under a Attribution-NonCommercial-NoDerivs creative commons License
No comments:
Post a Comment
Your comments are always welcome.