ログイン
Language:

WEKO3

  • トップ
  • ランキング
To
lat lon distance
To

Field does not validate



インデックスリンク

インデックスツリー

メールアドレスを入力してください。

WEKO

One fine body…

WEKO

One fine body…

アイテム

  1. 学位論文
  2. 学位論文

高精度線分検知を可能にする非線形連続系のハフ変換適用に関する研究

https://doi.org/10.18997/0002001052
https://doi.org/10.18997/0002001052
8fffb2ab-0d69-423b-a807-c68f4fffb3c8
名前 / ファイル ライセンス アクション
sei_k_471.pdf sei_k_471.pdf (13.8 MB)
アイテムタイプ 学位論文 = Thesis or Dissertation(1)
公開日 2024-11-20
資源タイプ
資源タイプ識別子 http://purl.org/coar/resource_type/c_db06
資源タイプ doctoral thesis
タイトル
タイトル A Nonlinear Continuous Dynamics in Dual Spaces Applied to the Hough Transform Computation for Detecting Linear Segments to Guarantee a High Degree of Accuracy
言語 en
タイトル
タイトル 高精度線分検知を可能にする非線形連続系のハフ変換適用に関する研究
言語 ja
言語
言語 jpn
著者 Purev Amarbold,

× Purev Amarbold,

en Purev Amarbold,

Search repository
抄録
内容記述タイプ Abstract
内容記述 The Standard Hough Transform (SHT) is a feature extraction technique that transforms edge points from image space to sinusoidal waves in parameter space. The intersections of these waves describe the features of a line in the image space. SHT uses an accumulation array to detect the intersections. The accumulation array divides the parameter space into a matrix, and whenever a sine wave goes through its cells, the value of that cell increments by one, which is called the voting procedure. Finally, after transforming edges and voting the cells of the accumulation array, the cell with the highest value describes a line in the image.
Experts widely recognize the SHT algorithm, extensively studying it for over six decades. Researchers have introduced numerous improvements and extensions to the algorithm in the past. Furthermore, all of them still utilize the accumulation array. The assumption of the accumulation array that the cell with the highest number of votes will describe the shape in the image is statistically valid if the data in the image space is sufficiently large.
The current work shows specific scenarios where the above assumption fails and provides theoretical proof. Accordingly, an improved intersection detection method in the parameter space is needed. The immediate objective of this thesis is to replace the accumulation array with attractor dynamics in Coupled Nonlinear Oscillator (CNO). Instead of dividing the parameter space into smaller cells, current work proposes to generate individual oscillators on each wave and synchronize them on the intersection points.
The thesis provides a theoretical framework for a model design using the Hough equation and inspects two coupling methods: pair-wise and global. Each coupling method has been tested on synthetic images of different configurations with heavy noise and compared based on detection accuracy. Pair-wise coupling method was found to be superior to the global.
In conclusion, a severe limitation of the SHT was noted as the mathematical proof if it uses an accumulation array for counting the number of points in each cell. Additionally, a nonlinear dynamics to replace the voting procedure was proposed, and the performance was successfully demonstrated in the computer experiment. Lastly, the proposed framework was verified to extend for line detection and other shapes if it has a conversion function that maps into a finite range. It contributes to the robust shape detection practice in field experiments.
目次
内容記述タイプ TableOfContents
内容記述 1 Introduction| 2 Literature Review| 3 Basic Concept| 4 Results| 5 Discussion| 6 Conclusion
備考
内容記述タイプ Other
内容記述 九州工業大学博士学位論文 学位記番号:生工博甲第471号 学位授与年月日:令和5年9月25日
学位授与番号
学位授与番号 甲第471号
学位名
学位名 博士(工学)
学位授与年月日
学位授与年月日 2023-09-25
学位授与機関
学位授与機関識別子Scheme kakenhi
学位授与機関識別子 17104
学位授与機関名 九州工業大学
言語 ja
学位授与年度
内容記述タイプ Other
内容記述 令和5年度
出版タイプ
出版タイプ VoR
出版タイプResource http://purl.org/coar/version/c_970fb48d4fbd8a85
アクセス権
アクセス権 open access
アクセス権URI http://purl.org/coar/access_right/c_abf2
ID登録
ID登録 10.18997/0002001052
ID登録タイプ JaLC
戻る
0
views
See details
Views

Versions

Ver.1 2024-11-20 06:29:13.900584
Show All versions

Share

Share
tweet

Cite as

Other

print

エクスポート

OAI-PMH
  • OAI-PMH JPCOAR 2.0
  • OAI-PMH JPCOAR 1.0
  • OAI-PMH DublinCore
  • OAI-PMH DDI
Other Formats
  • JSON
  • BIBTEX
  • ZIP

コミュニティ

確認

確認

確認


Powered by WEKO3


Powered by WEKO3