國立虎尾科技大學 |

Handwriting Verification Using Deep Convolutional Neural Networks.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Handwriting Verification Using Deep Convolutional Neural Networks./
作者:	Ramprasad, Sanjana.
面頁冊數:	1 online resource (92 pages)
附註:	Source: Masters Abstracts International, Volume: 56-05.
標題:	Computer science. -
電子資源:	click for full text (PQDT)
ISBN:	9780355046694

Handwriting Verification Using Deep Convolutional Neural Networks.
Ramprasad, Sanjana.

Handwriting Verification Using Deep Convolutional Neural Networks. - 1 online resource (92 pages)

Source: Masters Abstracts International, Volume: 56-05.

Thesis (M.S.)--State University of New York at Buffalo, 2017.

Includes bibliographical references

With the growing popularity Convolutional Neural Networks(CNNs) in computer vision tasks and the ability of them to model intricate patterns in data, we took to extending the study of verifying handwriting using CNNs. Handwriting has been considered to be unique for each individual, which is why they are still of considerable importance in legal documents. Although handwriting is considered to be unique, there still exists variations within a writer's handwriting. We thus seek to build a model that can successfully differentiate writers while remaining invariant to within writer changes without feature engineering as used by previous models. (Individuality of Handwriting, S. N. Srihari et.al 2002).

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355046694Subjects--Topical Terms:

573171
Computer science.
Index Terms--Genre/Form:

554714
Electronic books.

Handwriting Verification Using Deep Convolutional Neural Networks.
LDR:03610ntm a2200361K 4500 001 912159
005 20180608102940.5
006 m o u
007 cr mn||||a|a||
008 190606s2017 xx obm 000 0 eng d
020 $a 9780355046694
035 $a (MiAaPQ)AAI10281514
035 $a (MiAaPQ)buffalo:15105
035 $a AAI10281514
040 $a MiAaPQ $b eng $c MiAaPQ
100 1 $a Ramprasad, Sanjana. $3 1184390
245 1 0 $a Handwriting Verification Using Deep Convolutional Neural Networks.
264 0 $c 2017
300 $a 1 online resource (92 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 56-05.
500 $a Adviser: Sargur N. Srihari.
502 $a Thesis (M.S.)--State University of New York at Buffalo, 2017.
504 $a Includes bibliographical references
520 $a With the growing popularity Convolutional Neural Networks(CNNs) in computer vision tasks and the ability of them to model intricate patterns in data, we took to extending the study of verifying handwriting using CNNs. Handwriting has been considered to be unique for each individual, which is why they are still of considerable importance in legal documents. Although handwriting is considered to be unique, there still exists variations within a writer's handwriting. We thus seek to build a model that can successfully differentiate writers while remaining invariant to within writer changes without feature engineering as used by previous models. (Individuality of Handwriting, S. N. Srihari et.al 2002).
520 $a We use the CEDAR ,word-level "and" dataset to train and experiment with several CNN architectures based on differences in filter size, feature maps, activation functions, regularization techniques, cost functions, width and depth of the model, pooling operations, strides and optimization techniques. While experimenting with an architecture, we found the model behaved with certain trends, by modeling several of these trends we get an intuitive understanding of the effect of each of these parameters on the model's performance. We use this understanding in each experiment for developing our final architecture. We further split the final architecture into two CNN architectures with the same parameter tuning along with weight and variable sharing and saw a spike in performance.
520 $a The primary objective of our experiments was to test how handwriting verification performs without any feature engineering and using Convolutional Neural Networks for feature extraction. However, we also gain several insights into trends in different architecture results with respect to the dataset and develop a final combination of parameters that was found to perform the best. We hope these insights could be used while tuning more complex models.
520 $a We test the model's performance by identifying two types of error based on model writers: 1) Samples from a known writer---This includes for testing variation within writer's samples seen before as well as between writers 2) Samples from an unknown writer---This includes for testing the generalizing capability of the model on samples of never before seen writers. By measuring these two kinds of errors, we arrive at an architecture that performs optimally for both.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Artificial intelligence. $3 559380
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a State University of New York at Buffalo. $b Computer Science and Engineering. $3 1180201
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10281514 $z click for full text (PQDT)