Loss function for ambiguous boundaries for deep neural network (DNN) for image segmentation

被引：1

作者：

Hakumura, Yuma ^{[1
]}

Ito, Taiyo ^{[1
]}

Matsui, Shiori ^{[1
]}

Akiba, Yuya ^{[1
]}

Aoki, Kimiya ^{[1
]}

Nakashima, Yuki ^{[2
]}

Hirao, Kiyoshi ^{[2
]}

Fukushima, Manabu ^{[2
]}

机构：

[1] Chukyo Univ, 101-2 Yagotohonmachi,Showa Ku, Nagoya, Aichi 4668666, Japan

[2] Natl Inst Adv Ind Sci & Technol, Multimat Res Inst, Ceram Microstruct Control Grp, Nagoya, Aichi, Japan

来源：

ELECTRONICS AND COMMUNICATIONS IN JAPAN | 2023年 / 106卷 / 04期

关键词：

blurriness of grain boundaries; deep neural network; fine ceramics; loss function; segmentation;

D O I：

10.1002/ecj.12429

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

This study deals with the task of segmentation of SEM images of fine ceramics sintered bodies by using deep neural network (DNN). In particular, we focus on misclassification caused by the blurriness of grain boundaries(boundaries between particles). Therefore, we utilize the frequency distribution of brightness gradient of grain boundaries and give higher weights to pixels with lower gradient values. Experiments confirmed that the model trained with proposed loss function gave the best prediction results.

引用

页数：9

共 50 条

[1] Development of a Psychoacoustic Loss Function for the Deep Neural Network (DNN)-Based Speech Coder
Byun, Joon
Shin, Seungmin
Park, Youngcheol
Sung, Jongmo
Bead, Seungkwon
INTERSPEECH 2021, 2021, : 1694 - 1698
[2] DEEP CONDITIONAL NEURAL NETWORK FOR IMAGE SEGMENTATION
Wang, Qiurui
Yuan, Chun
Liu, Yan
2016 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA & EXPO (ICME), 2016,
[3] Loss Function Selection in a Problem of Satellite Image Segmentation Using Convolutional Neural Network
Sedov, A. G.
Khryashchev, V. V.
Larionov, R. V.
Ostrovskaya, A. A.
2019 SYSTEMS OF SIGNAL SYNCHRONIZATION, GENERATING AND PROCESSING IN TELECOMMUNICATIONS (SYNCHROINFO), 2019,
[4] Learning Deep Conditional Neural Network for Image Segmentation
Wang, Qiurui
Yuan, Chun
Liu, Yan
IEEE TRANSACTIONS ON MULTIMEDIA, 2019, 21 (07) : 1839 - 1852
[5] A new imbalance-aware loss function to be used in a deep neural network for colorectal polyp segmentation
Gokkan, Ozan
Kuntalp, Mehmet
COMPUTERS IN BIOLOGY AND MEDICINE, 2022, 151
[6] Neural network for image segmentation
Skourikhine, AN
Prasad, L
Schlei, BR
APPLICATIONS AND SCIENCE OF NEURAL NETWORKS, FUZZY SYSTEMS, AND EVOLUTIONARY COMPUTATION III, 2000, 4120 : 28 - 35
[7] Deep Neural Networks with Intersection over Union Loss for Binary Image Segmentation
van Beers, Floris
Lindstrom, Arvid
Okafor, Emmanuel
Wiering, Marco A.
ICPRAM: PROCEEDINGS OF THE 8TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION APPLICATIONS AND METHODS, 2019, : 438 - 445
[8] A deep ensemble medical image segmentation with novel sampling method and loss function
Roshan S.
Tanha J.
Zarrin M.
Babaei A.F.
Nikkhah H.
Jafari Z.
Computers in Biology and Medicine, 2024, 172
[9] Deep Neural Network for Lung Image Segmentation on Chest X-ray
Chavan, Mahesh
Varadarajan, Vijayakumar
Gite, Shilpa
Kotecha, Ketan
TECHNOLOGIES, 2022, 10 (05)
[10] An AutoAssociative Neural Network for Image Segmentation
Marcolino dos Santos, Hugo Leonardo
Torres Fernandes, Bruno Jose
Maciel Fernandes, Sergio Murilo
2016 IEEE LATIN AMERICAN CONFERENCE ON COMPUTATIONAL INTELLIGENCE (LA-CCI), 2016,

← 1 2 3 4 5 →