A statistical approach to improve CNN classification accuracy

被引：0

作者：

Pentsos, Vasileios ^{[1
]}

Tragoudas, Spyros ^{[1
]}

机构：

[1] Southern Illinois Univ, Sch Elect Comp & Biomed Engn, Carbondale, IL 62901 USA

来源：

2023 IEEE 24TH INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE SWITCHING AND ROUTING, HPSR | 2023年

关键词：

D O I：

10.1109/HPSR57248.2023.10148033

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Convolutional neural networks (CNNs) have achieved state-of-the-art performance in image classification tasks. However, they may underperform for specific classes, resulting in misclassifications. To address this issue, the proposed method involves two steps that use the Mann-Whitney U test on generated image distributions. The method is evaluated on the publicly available dataset CIFAR100, utilizing ResNet-50 as the baseline network. The results show that the proposed method is effective in significantly improving the classification accuracy of low-accuracy image classes while preserving the high-accuracy classes.

引用

页数：5

共 50 条

[41] Can multiresolution fusion techniques improve classification accuracy?
Bruzzone, L.
Carlin, L.
Alparone, L.
Baronti, S.
Garzelli, A.
Nencini, F.
IMAGE AND SIGNAL PROCESSING FOR REMOTE SENSING XII, 2006, 6365
[42] CNN-BD: An Approach for Disease Classification and Visualization
Rao, G. Madhukar
Kumar, T. Ravi
Reddy, A. Rajashekar
ADVANCES IN DATA SCIENCE AND MANAGEMENT, 2020, 37 : 149 - 157
[43] A Generic Approach of ECG Signals Classification Using CNN
Naidu, P. Rama Santosh
Devi, G. Lavanya
HELIX, 2020, 10 (05): : 75 - 80
[44] Can Synthetic Images Improve CNN Performance in Wound Image Classification?
Malihi, Leila
Huebner, Ursula
Richter, Mats L.
Moelleken, Maurice
Przysucha, Mareike
Busch, Dorothee
Heggemann, Jan
Hafer, Guido
Wiemeyer, Stefan
Heidemann, Gunther
Dissemond, Joachim
Erfurt-Berge, Cornelia
Barkhau, Carlotta
Hendriks, Achim
Huesers, Jens
CARING IS SHARING-EXPLOITING THE VALUE IN DATA FOR HEALTH AND INNOVATION-PROCEEDINGS OF MIE 2023, 2023, 302 : 927 - 931
[45] Integrate Traditional Hand-Crafted Features into Modern CNN-based Models to Further Improve SAR Ship Classification Accuracy
Zhang, Tianwen
Zhang, Xiaoling
2021 7TH ASIA-PACIFIC CONFERENCE ON SYNTHETIC APERTURE RADAR (APSAR), 2021,
[46] Classification of flotation frothers - A statistical approach
Moreno, Yesenia Saavedra
Bournival, Ghislain
Ata, Seher
CHEMICAL ENGINEERING SCIENCE, 2022, 248
[47] CLASSIFICATION OF COSMIC SOURCES - A STATISTICAL APPROACH
GIOVANNELLI, F
CORADINI, A
LASOTA, JP
POLIMENE, ML
ASTRONOMY & ASTROPHYSICS, 1981, 95 (01) : 138 - 142
[48] A STATISTICAL APPROACH FOR SIMULTANEOUS SEGMENTATION AND CLASSIFICATION
Zanotta, Daniel C.
Ferreira, Matheus P.
Zorte, Maciel
Shimabukuro, Yosio
2014 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS), 2014, : 4899 - 4901
[49] APHASIA CLASSIFICATION BY MULTIDIMENSIONAL STATISTICAL APPROACH
MATVEEV, M
ELECTROENCEPHALOGRAPHY AND CLINICAL NEUROPHYSIOLOGY, 1980, 50 (3-4): : P102 - P102
[50] A statistical approach to TCP session classification
Moscalu, Tudor
Steel, Andrew M.
Brown, Edward D. L.
Lim, Yangkind L.
2008 SYSTEMS AND INFORMATION ENGINEERING DESIGN SYMPOSIUM, 2008, : 11 - 16

← 1 2 3 4 5 →