BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:US/Eastern
BEGIN:DAYLIGHT
DTSTART:20170312T030000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20171105T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20170502T203949Z
UID:133A7E7F-081A-11E7-A7C6-0050568D7F66
DTSTART;TZID=US/Eastern:20170505T113000
DTEND;TZID=US/Eastern:20170505T130000
DESCRIPTION:Much research efforts have been devoted to applying machine lea
 rning algorithms in video imagery for object recognition. However\, very l
 imited open literatures can be found on machine learning in radio frequenc
 y data. Hence\, this research explores application of deep learning algori
 thms to the task of Automatic Target Recognition (ATR) in synthetic apertu
 re radar (SAR) imagery. Radar enables imaging ground objects at far greate
 r standoff distances. However\, the false-alarm rate of both human and mac
 hine-based radar image recognition is unacceptably high. Existing ATR algo
 rithms also require impractically large computing resources for airborne a
 pplications. The goal of our research is to advance the state-of-the-art A
 TR capability by developing a more accurate\, real-time\, and low-power ob
 ject recognition system. We implemented Convolution Neural Network (CNN) b
 ased SAR object recognition algorithms in GPU and energy efficient computi
 ng systems. We received acceptable classification accuracy on relevant SAR
  data. We will discuss technical challenges and future research on radio f
 requency object recognition.\n\nSpeaker(s): Dr. Uttam Majumder\, \, Dr. Ut
 tam Majumder\, \n\nAgenda: \n11:30 AM Social/Pizza\n\n12:00 PM Presentatio
 n\n\nRoom: 1435\, Bldg: Golisano Hall (GOL)\, 152 Lamb Memorial Dr.\, Roch
 ester\, New York\, United States\, 14623
LOCATION:Room: 1435\, Bldg: Golisano Hall (GOL)\, 152 Lamb Memorial Dr.\, R
 ochester\, New York\, United States\, 14623
ORGANIZER:bo.yuan@rit.edu
SEQUENCE:7
SUMMARY:Machine Learning for Object Recognition from High Volume Radio Freq
 uency Data
URL;VALUE=URI:https://events.vtools.ieee.org/m/44454
X-ALT-DESC:Description: <br /><p>Much research efforts have been devoted to
  applying machine learning algorithms in video imagery for object recognit
 ion. However\, very limited open literatures can be found on machine learn
 ing in radio frequency data. Hence\, this research explores application of
  deep learning algorithms to the task of Automatic Target Recognition (ATR
 ) in synthetic aperture radar (SAR) imagery. Radar enables imaging ground 
 objects at far greater standoff distances. However\, the false-alarm rate 
 of both human and machine-based radar image recognition is unacceptably hi
 gh. Existing ATR algorithms also require impractically large computing res
 ources for airborne applications. The goal of our research is to advance t
 he state-of-the-art ATR capability by developing a more accurate\, real-ti
 me\, and low-power object recognition system. We implemented Convolution N
 eural Network (CNN) based SAR object recognition algorithms in GPU and ene
 rgy efficient computing systems. We received acceptable classification acc
 uracy on relevant SAR data. We will discuss technical challenges and futur
 e research on radio frequency object recognition.</p><br /><br />Agenda: <
 br /><p>11:30 AM Social/Pizza</p>\n<p>12:00 PM Presentation</p>
END:VEVENT
END:VCALENDAR