東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Real-time active robotic vision usin...

Garaas, Tyler W.

FindBook

Google Book

Amazon

博客來

Real-time active robotic vision using biologically inspired neural models.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Real-time active robotic vision using biologically inspired neural models./
作者:	Garaas, Tyler W.
面頁冊數:	166 p.
附註:	Source: Dissertation Abstracts International, Volume: 71-09, Section: B, page: 5576.
Contained By:	Dissertation Abstracts International71-09B.
標題:	Biology, Neuroscience. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3420079
ISBN:	9781124153667

Real-time active robotic vision using biologically inspired neural models.
Garaas, Tyler W.

Real-time active robotic vision using biologically inspired neural models. - 166 p.

Source: Dissertation Abstracts International, Volume: 71-09, Section: B, page: 5576.

Thesis (Ph.D.)--University of Massachusetts Boston, 2010.

The discipline known as computer vision strives to extract a sense of meaning from the patterns of light falling across a sensor. Many difficulties hinder the advancement of this cause, ranging from ambiguity as to what form best encapsulates the structure of the information, to the intractable computational bandwidth required to process sensor images of sufficiently high resolution. One promising outlet to overcoming such difficulties is to learn from the single system that has already found solutions to many of the inherent problems: biological vision.

ISBN: 9781124153667Subjects--Topical Terms:

1017680
Biology, Neuroscience.

Real-time active robotic vision using biologically inspired neural models.
LDR:03138nam 2200337 4500 001 1405127
005 20111206130411.5
008 130515s2010 ||||||||||||||||| ||eng d
020 $a 9781124153667
035 $a (UMI)AAI3420079
035 $a AAI3420079
040 $a UMI $c UMI
100 1 $a Garaas, Tyler W. $3 1684482
245 1 0 $a Real-time active robotic vision using biologically inspired neural models.
300 $a 166 p.
500 $a Source: Dissertation Abstracts International, Volume: 71-09, Section: B, page: 5576.
500 $a Adviser: Marc Pomplun.
502 $a Thesis (Ph.D.)--University of Massachusetts Boston, 2010.
520 $a The discipline known as computer vision strives to extract a sense of meaning from the patterns of light falling across a sensor. Many difficulties hinder the advancement of this cause, ranging from ambiguity as to what form best encapsulates the structure of the information, to the intractable computational bandwidth required to process sensor images of sufficiently high resolution. One promising outlet to overcoming such difficulties is to learn from the single system that has already found solutions to many of the inherent problems: biological vision.
520 $a In the present work, I propose a system of neurons inspired by a subset of the visual areas of the primate brain in order to direct the "gaze" of a pan/tilt/zoom camera in real-time to areas of interest that are demarcated by one or more low-level visual features known to form the basis of visual abilities in primates: color, form, and motion. The ultimate goal of work in this area is to achieve a system that can visually navigate a scene in a similar manner and time-frame as humans.
520 $a In order to overcome the high computational bandwidth required to model hundreds-of-thousands of neurons connected by millions of synapses in real-time, I exploit the highly parallel nature of the neural model to develop a computational model on recently available parallel computing architecture known as the graphics processing unit (GPU).
520 $a The proposed neural model includes areas to control various aspects of the physical camera system. In particular, I present novel methods for controlling the white- balance, exposure, and orientation of the camera system.
520 $a Finally, the ability of the robotic vision system to effectively direct overt visual attention is contrasted with that of human participants. From the experiment, attentional landscapes were computed, which demonstrated that deployment of overt visual attention by human participants and the robotic system was similar. Furthermore, results from the model were comparable with previous neural models that are neither robotic nor executed in real-time. As such, the current work could form the basis of a future sophisticated robotic vision system.
590 $a School code: 1074.
650 4 $a Biology, Neuroscience. $3 1017680
650 4 $a Engineering, Robotics. $3 1018454
650 4 $a Computer Science. $3 626642
690 $a 0317
690 $a 0771
690 $a 0984
710 2 $a University of Massachusetts Boston. $3 1024079
773 0 $t Dissertation Abstracts International $g 71-09B.
790 1 0 $a Pomplun, Marc, $e advisor
790 $a 1074
791 $a Ph.D.
792 $a 2010
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3420079