BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:US/Mountain
BEGIN:DAYLIGHT
DTSTART:20140309T030000
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:MDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20141102T010000
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:MST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20140602T032239Z
UID:B787811A-3B4E-1032-825E-0050568D3657
DTSTART;TZID=US/Mountain:20140613T180000
DTEND;TZID=US/Mountain:20140613T203000
DESCRIPTION:Topic: Wildfires are a constant and present danger in our commu
 nity with early detection being an important component in wildfire eradica
 tion. Please join us in Room W-101 on the Rampart Range Campus of Pikes Pe
 ak Community College June 13 at 6:00 to hear a presentation by Dr. Trung P
 ham on automatic detection of wildfire through spectral analysis of aerial
  images. Abstract: This paper proposes extending the threshold filtering t
 echnique on color images of three-dimensional RGB pixels to detect fire in
  order to take advantage of the computational efficiency of the technique.
  In this extension\, the three-dimensional color spectrum is analyzed for 
 detection of fire\, and set theory is used to analytically describe a boun
 ded region in a three-dimensional space. Additionally\, the K-Mean cluster
 ing technique normally used effectively for segmentation is extended from 
 identifying a cluster by it center point to identifying a cluster’s
  center with a set to allow flexibility in the initialization process and 
 faster converging rate when initial condition was accurately assessed. A s
 pecific structure of an image into three segments of flame\, smoke\, and e
 verything else is formulated mathematically to eliminate the need of prior
  establishing the condition of fire normally required for the K-Mean segme
 ntation. Together\, the combination of multi-dimensional threshold filter\
 , K-Mean clustering based on set theory\, and three-part segment structuri
 ng offers closed form analytical is effective in detection of fire\, effic
 ient in computational requirement\, and capable of handling extreme cases 
 normally not feasible with traditional K-Mean segmentation. Abstract funct
 ional analysis is provided to demonstrate the convergence of the segmentat
 ion process. Numerical examples are given to illustrate the concept of fir
 e detection through spectral analysis of aerial images\, and to demonstrat
 e the workability of the solution. Meeting Details: Attendance is free to 
 both IEEE members and non-members. Pizza and beverages (soda and water) wi
 ll be provided. A $5.00 donation from non-members would be greatly appreci
 ated to defer costs. Please register at the below URL as seating is limite
 d. Simply click on the link to be taken to the registration web page. Ques
 tions should be addressed to Russ Bogardus at russbogardus@comcast.net or 
 (719) 337-6328. Driving Directions: The Rampart Range Campus is located on
  the north end of Colorado Springs and on the east side of Interstate 25 a
 nd InterQuest Parkway\, exit 153. Coming from either the north or south on
  I-25\, take the Interquest Parkway exit (#153) and proceed east. Proceed 
 to the 4th stoplight and turn left on New Life Drive. Take another quick l
 eft and turn into the campus. Detailed maps are given at: http://www.ppcc.
 edu/about/campus-mapsdirections\n\nCo-sponsored by: IEEE PPS Computer Soci
 ety\n\nSpeaker(s): Dr. Trung Pham\, \n\nAgenda: \nJune 13 6:00 - 6:30 Meet
 ing meet and greet. Pizza and beverages available. 6:30 - 8:00 Presentatio
 n 8:00 - 8:30 Dr. Pham available for questions\n\nRoom: W-101\, Bldg: Main
  Building - Lower Level\, 11195 Highway 83\, Colorado Springs\, Colorado\,
  United States\, 80921
LOCATION:Room: W-101\, Bldg: Main Building - Lower Level\, 11195 Highway 83
 \, Colorado Springs\, Colorado\, United States\, 80921
ORGANIZER:russbogardus@comcast.net
SEQUENCE:5
SUMMARY:IEEE Pikes Peak Section Presentation: Automatic Detection of Wildfi
 re Through Spectral Analysis of Aerial Images
URL;VALUE=URI:https://events.vtools.ieee.org/m/26637
X-ALT-DESC:Description: <br />Topic: Wildfires are a constant and present d
 anger in our community with early detection being an important component i
 n wildfire eradication. Please join us in Room W-101 on the Rampart Range 
 Campus of Pikes Peak Community College June 13 at 6:00 to hear a presentat
 ion by Dr. Trung Pham on automatic detection of wildfire through spectral 
 analysis of aerial images.\n \n\nAbstract: This paper proposes extending t
 he threshold filtering technique on color images of three-dimensional RGB 
 pixels to detect fire in order to take advantage of the computational effi
 ciency of the technique. In this extension\, the three-dimensional color s
 pectrum is analyzed for detection of fire\, and set theory is used to anal
 ytically describe a bounded region in a three-dimensional space. Additiona
 lly\, the K-Mean clustering technique normally used effectively for segmen
 tation is extended from identifying a cluster by it center point to identi
 fying a cluster’s center with a set to allow flexibility in the ini
 tialization process and faster converging rate when initial condition was 
 accurately assessed. A specific structure of an image into three segments 
 of flame\, smoke\, and everything else is formulated mathematically to eli
 minate the need of prior establishing the condition of fire normally requi
 red for the K-Mean segmentation. Together\, the combination of multi-dimen
 sional threshold filter\, K-Mean clustering based on set theory\, and thre
 e-part segment structuring offers closed form analytical is effective in d
 etection of fire\, efficient in computational requirement\, and capable of
  handling extreme cases normally not feasible with traditional K-Mean segm
 entation. Abstract functional analysis is provided to demonstrate the conv
 ergence of the segmentation process. Numerical examples are given to illus
 trate the concept of fire detection through spectral analysis of aerial im
 ages\, and to demonstrate the workability of the solution.\n\n\nMeeting De
 tails: Attendance is free to both IEEE members and non-members. Pizza and 
 beverages (soda and water) will be provided. A $5.00 donation from non-mem
 bers would be greatly appreciated to defer costs. Please register at the b
 elow URL as seating is limited. Simply click on the link to be taken to th
 e registration web page. Questions should be addressed to Russ Bogardus at
  russbogardus@comcast.net or (719) 337-6328.\n\n\nDriving Directions: The 
 Rampart Range Campus is located on the north end of Colorado Springs and o
 n the east side of Interstate 25 and InterQuest Parkway\, exit 153.  Comin
 g from either the north or south on I-25\, take the Interquest Parkway exi
 t (#153) and proceed east. Proceed to the 4th stoplight and turn left on N
 ew Life Drive. Take another quick left and turn into the campus. Detailed 
 maps are given at: http://www.ppcc.edu/about/campus-mapsdirections<br /><b
 r />Agenda: <br />June 13\n\n\n6:00 - 6:30    Meeting meet and greet. Pizz
 a and beverages available.\n\n6:30 - 8:00    Presentation\n\n8:00 - 8:30  
   Dr. Pham available for questions 
END:VEVENT
END:VCALENDAR