My Math Forum  

Go Back   My Math Forum > College Math Forum > Advanced Statistics

Advanced Statistics Advanced Probability and Statistics Math Forum

LinkBack Thread Tools Display Modes
October 14th, 2008, 07:55 AM   #1
Joined: Oct 2008

Posts: 2
Thanks: 0

Courses needed for biometric analysis

Hi. I donít have a lot of math experience, but Iíd like to get some advice about what types of math I would need to learn to find out if there is a statistical relationship between changes in the geographical distribution of gait patterns and other geographical factors. So far, I have broken down this question into the following questions. What math would I need to analyze computer language that was interpreted from a video camera in order to recognize human gait from the background that would also be picked up by the video camera? What math would I need to efficiently represent the human gait that was recognized in order to classify similar gait types? What math would I need to represent the distribution of gait types in relation to each other and geographically? What math would I need to classify geographical regions in terms of the distribution of these gait types? Letís say that I identified w different geographical regions that communicate with each other; and each geographical region has x characteristics; and there are y different communication distances between the w different geographical regions; and z types of communications. Each type of communication could either involve forwarding, returning, initiating, or increasing existing communication type and route. What math would I need to also classify geographical regions in terms of these communication patterns? If similar later changes in the above variables are hypothesized to be due to similar earlier changes in the above variables, what math would I need to identify what types of later changes tend to follow what types of earlier changes?
JasonB is offline  
October 14th, 2008, 08:05 AM   #2
Global Moderator
CRGreathouse's Avatar
Joined: Nov 2006
From: UTC -5

Posts: 16,046
Thanks: 938

Math Focus: Number theory, computational mathematics, combinatorics, FOM, symbolic logic, TCS, algorithms
Re: Courses needed for biometric analysis

The hardest part, by far, is recognizing the gait. This is pattern recognition, which I think of more as computer science than mathematics per se, but they're closely related in any case. The analysis of gait vs. geography is statistics, and shouldn't take more than a basic understanding (a college course or two in stats should suffice).
CRGreathouse is offline  
October 14th, 2008, 09:06 AM   #3
Senior Member
Joined: Oct 2007
From: Chicago

Posts: 1,701
Thanks: 3

Re: Courses needed for biometric analysis

Actually analyzing the gaits would definitely be a CS thing. A lot of AI-- some rather advanced probability (still undergraduate, but tricky...), and a decent understanding of data structures, sorting and searching.
Anyway, the prob/stats class CRG recommended + "intro to discrete" + (either) "Artificial intelligence" (or) "Algorithms and data structures"...
There may be another prerequisite for either of the last two. I would recommend taking an AI class (specifically one on visual recognition, if you can find it), over an algorithms class, as it will be much more applicable.

Other than that... I don't know. Find an AI researcher to work with you? I'm sure you can find one who is interested?
cknapp is offline  

  My Math Forum > College Math Forum > Advanced Statistics

analysis, biometric, courses, needed

Thread Tools
Display Modes

Similar Threads
Thread Thread Starter Forum Replies Last Post
Courses in Trigonometry Question shiseonji Trigonometry 1 July 28th, 2013 05:58 AM
About online courses or anything else numproc Academic Guidance 1 July 22nd, 2013 06:12 PM
Cyber Forensics Courses homertyson Computer Science 0 March 6th, 2013 08:21 PM
Testing out of courses JasonB Academic Guidance 1 January 17th, 2010 06:16 PM
Best books for theses courses sivela Academic Guidance 2 December 31st, 1969 04:00 PM

Copyright © 2019 My Math Forum. All rights reserved.