The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Cet article propose principalement une méthode de détection de segments de ligne basée sur la suppression de pseudo-pics et la transformée de Hough locale, qui présente une bonne résistance au bruit et peut résoudre les problèmes de détection manquante de segments de ligne courts, de fausse détection et de sursegmentation. De plus, en réponse au phénomène de développement inégal dans les images tomographiques en émulsion nucléaire, cet article propose un processus de prétraitement d'image qui utilise la méthode « Différence de Gauss » pour réduire le bruit, puis utilise l'écart type de la valeur de gris de chaque pixel pour regrouper et unifier la valeur de gris de chaque pixel, ce qui permet d'obtenir de manière robuste les caractéristiques linéaires de ces images. Les tests sur l'ensemble de données réel d'images tomographiques d'émulsion nucléaire et sur l'ensemble de données public YorkUrban montrent que la méthode proposée peut améliorer efficacement la précision du réseau neuronal convolutif ou de la vision dans la classification des événements basée sur le transformateur pour les événements de désintégration alpha dans l'émulsion nucléaire. En particulier, la méthode de détection de segments de ligne dans la méthode proposée permet d'obtenir des résultats optimaux en termes de précision et de vitesse de traitement, ce qui présente également une forte capacité de généralisation dans des images naturelles de haute qualité.
Ye TIAN
Zhuzhou CRRC Times Electric Co., Ltd.
Mei HAN
Hunan University of Technology
Jinyi ZHANG
Shenyang Ligong University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Ye TIAN, Mei HAN, Jinyi ZHANG, "Line Segment Detection Based on False Peak Suppression and Local Hough Transform and Application to Nuclear Emulsion" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 11, pp. 1854-1867, November 2023, doi: 10.1587/transinf.2023EDP7117.
Abstract: This paper mainly proposes a line segment detection method based on pseudo peak suppression and local Hough transform, which has good noise resistance and can solve the problems of short line segment missing detection, false detection, and oversegmentation. In addition, in response to the phenomenon of uneven development in nuclear emulsion tomographic images, this paper proposes an image preprocessing process that uses the “Difference of Gaussian” method to reduce noise and then uses the standard deviation of the gray value of each pixel to bundle and unify the gray value of each pixel, which can robustly obtain the linear features in these images. The tests on the actual dataset of nuclear emulsion tomographic images and the public YorkUrban dataset show that the proposed method can effectively improve the accuracy of convolutional neural network or vision in transformer-based event classification for alpha-decay events in nuclear emulsion. In particular, the line segment detection method in the proposed method achieves optimal results in both accuracy and processing speed, which also has strong generalization ability in high quality natural images.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2023EDP7117/_p
Copier
@ARTICLE{e106-d_11_1854,
author={Ye TIAN, Mei HAN, Jinyi ZHANG, },
journal={IEICE TRANSACTIONS on Information},
title={Line Segment Detection Based on False Peak Suppression and Local Hough Transform and Application to Nuclear Emulsion},
year={2023},
volume={E106-D},
number={11},
pages={1854-1867},
abstract={This paper mainly proposes a line segment detection method based on pseudo peak suppression and local Hough transform, which has good noise resistance and can solve the problems of short line segment missing detection, false detection, and oversegmentation. In addition, in response to the phenomenon of uneven development in nuclear emulsion tomographic images, this paper proposes an image preprocessing process that uses the “Difference of Gaussian” method to reduce noise and then uses the standard deviation of the gray value of each pixel to bundle and unify the gray value of each pixel, which can robustly obtain the linear features in these images. The tests on the actual dataset of nuclear emulsion tomographic images and the public YorkUrban dataset show that the proposed method can effectively improve the accuracy of convolutional neural network or vision in transformer-based event classification for alpha-decay events in nuclear emulsion. In particular, the line segment detection method in the proposed method achieves optimal results in both accuracy and processing speed, which also has strong generalization ability in high quality natural images.},
keywords={},
doi={10.1587/transinf.2023EDP7117},
ISSN={1745-1361},
month={November},}
Copier
TY - JOUR
TI - Line Segment Detection Based on False Peak Suppression and Local Hough Transform and Application to Nuclear Emulsion
T2 - IEICE TRANSACTIONS on Information
SP - 1854
EP - 1867
AU - Ye TIAN
AU - Mei HAN
AU - Jinyi ZHANG
PY - 2023
DO - 10.1587/transinf.2023EDP7117
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2023
AB - This paper mainly proposes a line segment detection method based on pseudo peak suppression and local Hough transform, which has good noise resistance and can solve the problems of short line segment missing detection, false detection, and oversegmentation. In addition, in response to the phenomenon of uneven development in nuclear emulsion tomographic images, this paper proposes an image preprocessing process that uses the “Difference of Gaussian” method to reduce noise and then uses the standard deviation of the gray value of each pixel to bundle and unify the gray value of each pixel, which can robustly obtain the linear features in these images. The tests on the actual dataset of nuclear emulsion tomographic images and the public YorkUrban dataset show that the proposed method can effectively improve the accuracy of convolutional neural network or vision in transformer-based event classification for alpha-decay events in nuclear emulsion. In particular, the line segment detection method in the proposed method achieves optimal results in both accuracy and processing speed, which also has strong generalization ability in high quality natural images.
ER -