Fruit Ripeness Classification with Few-Shot Learning

Hui Fuang Ng; Jie Jin Lo; Chih Yang Lin; Hung Khoon Tan; Joon Huang Chuah; Kar Hang Leung

doi:10.1007/978-981-16-8129-5_109

Fruit Ripeness Classification with Few-Shot Learning

Hui Fuang Ng, Jie Jin Lo, Chih Yang Lin, Hung Khoon Tan, Joon Huang Chuah, Kar Hang Leung

Department of Mechanical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

Deep learning based image classification systems require large amount of training data and long training time. However, the availability of large annotated image dataset is usually limited and expensive to generate, which limits a vision system to adapt to new task efficiently. In this paper, a few-shot classification framework is proposed which can adapt one fruit ripeness classification system to classify new types of fruits using only a few training samples. The proposed framework adopts the meta-learning paradigm where a base network learns to extract meta-features and few-shot classification tasks from the base classes with abundant training samples and then apply the network to similar task on the novel classes using only a few support samples. Experimental results indicate that the proposed framework is able to achieve over 75% ripeness classification accuracy on various fruits using a little as five samples.

Original language	English
Title of host publication	Proceedings of the 11th International Conference on Robotics, Vision, Signal Processing and Power Applications - Enhancing Research and Innovation through the Fourth Industrial Revolution
Editors	Nor Muzlifah Mahyuddin, Nor Rizuan Mat Noor, Harsa Amylia Mat Sakim
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	715-720
Number of pages	6
ISBN (Print)	9789811681288
DOIs	https://doi.org/10.1007/978-981-16-8129-5_109
State	Published - 2022
Event	11th International Conference on Robotics, Vision, Signal Processing and Power Applications, RoViSP 2021 - Virtual, Online Duration: 5 Apr 2021 → 6 Apr 2021

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	829 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	11th International Conference on Robotics, Vision, Signal Processing and Power Applications, RoViSP 2021
City	Virtual, Online
Period	5/04/21 → 6/04/21

Keywords

Classification
Deep learning
Few-shot learning
Fruit ripeness

Access to Document

10.1007/978-981-16-8129-5_109

Cite this

Ng, H. F., Lo, J. J., Lin, C. Y., Tan, H. K., Chuah, J. H., & Leung, K. H. (2022). Fruit Ripeness Classification with Few-Shot Learning. In N. M. Mahyuddin, N. R. Mat Noor, & H. A. Mat Sakim (Eds.), Proceedings of the 11th International Conference on Robotics, Vision, Signal Processing and Power Applications - Enhancing Research and Innovation through the Fourth Industrial Revolution (pp. 715-720). (Lecture Notes in Electrical Engineering; Vol. 829 LNEE). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-16-8129-5_109

Ng, Hui Fuang ; Lo, Jie Jin ; Lin, Chih Yang et al. / Fruit Ripeness Classification with Few-Shot Learning. Proceedings of the 11th International Conference on Robotics, Vision, Signal Processing and Power Applications - Enhancing Research and Innovation through the Fourth Industrial Revolution. editor / Nor Muzlifah Mahyuddin ; Nor Rizuan Mat Noor ; Harsa Amylia Mat Sakim. Springer Science and Business Media Deutschland GmbH, 2022. pp. 715-720 (Lecture Notes in Electrical Engineering).

@inproceedings{7a80d79e7df749c08233defb75bb75e6,

title = "Fruit Ripeness Classification with Few-Shot Learning",

abstract = "Deep learning based image classification systems require large amount of training data and long training time. However, the availability of large annotated image dataset is usually limited and expensive to generate, which limits a vision system to adapt to new task efficiently. In this paper, a few-shot classification framework is proposed which can adapt one fruit ripeness classification system to classify new types of fruits using only a few training samples. The proposed framework adopts the meta-learning paradigm where a base network learns to extract meta-features and few-shot classification tasks from the base classes with abundant training samples and then apply the network to similar task on the novel classes using only a few support samples. Experimental results indicate that the proposed framework is able to achieve over 75% ripeness classification accuracy on various fruits using a little as five samples.",

keywords = "Classification, Deep learning, Few-shot learning, Fruit ripeness",

author = "Ng, {Hui Fuang} and Lo, {Jie Jin} and Lin, {Chih Yang} and Tan, {Hung Khoon} and Chuah, {Joon Huang} and Leung, {Kar Hang}",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; 11th International Conference on Robotics, Vision, Signal Processing and Power Applications, RoViSP 2021 ; Conference date: 05-04-2021 Through 06-04-2021",

year = "2022",

doi = "10.1007/978-981-16-8129-5_109",

language = "???core.languages.en_GB???",

isbn = "9789811681288",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "715--720",

editor = "Mahyuddin, {Nor Muzlifah} and {Mat Noor}, {Nor Rizuan} and {Mat Sakim}, {Harsa Amylia}",

booktitle = "Proceedings of the 11th International Conference on Robotics, Vision, Signal Processing and Power Applications - Enhancing Research and Innovation through the Fourth Industrial Revolution",

}

Ng, HF, Lo, JJ, Lin, CY, Tan, HK, Chuah, JH & Leung, KH 2022, Fruit Ripeness Classification with Few-Shot Learning. in NM Mahyuddin, NR Mat Noor & HA Mat Sakim (eds), Proceedings of the 11th International Conference on Robotics, Vision, Signal Processing and Power Applications - Enhancing Research and Innovation through the Fourth Industrial Revolution. Lecture Notes in Electrical Engineering, vol. 829 LNEE, Springer Science and Business Media Deutschland GmbH, pp. 715-720, 11th International Conference on Robotics, Vision, Signal Processing and Power Applications, RoViSP 2021, Virtual, Online, 5/04/21. https://doi.org/10.1007/978-981-16-8129-5_109

Fruit Ripeness Classification with Few-Shot Learning. / Ng, Hui Fuang; Lo, Jie Jin; Lin, Chih Yang et al.
Proceedings of the 11th International Conference on Robotics, Vision, Signal Processing and Power Applications - Enhancing Research and Innovation through the Fourth Industrial Revolution. ed. / Nor Muzlifah Mahyuddin; Nor Rizuan Mat Noor; Harsa Amylia Mat Sakim. Springer Science and Business Media Deutschland GmbH, 2022. p. 715-720 (Lecture Notes in Electrical Engineering; Vol. 829 LNEE).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Fruit Ripeness Classification with Few-Shot Learning

AU - Ng, Hui Fuang

AU - Lo, Jie Jin

AU - Lin, Chih Yang

AU - Tan, Hung Khoon

AU - Chuah, Joon Huang

AU - Leung, Kar Hang

PY - 2022

Y1 - 2022

N2 - Deep learning based image classification systems require large amount of training data and long training time. However, the availability of large annotated image dataset is usually limited and expensive to generate, which limits a vision system to adapt to new task efficiently. In this paper, a few-shot classification framework is proposed which can adapt one fruit ripeness classification system to classify new types of fruits using only a few training samples. The proposed framework adopts the meta-learning paradigm where a base network learns to extract meta-features and few-shot classification tasks from the base classes with abundant training samples and then apply the network to similar task on the novel classes using only a few support samples. Experimental results indicate that the proposed framework is able to achieve over 75% ripeness classification accuracy on various fruits using a little as five samples.

AB - Deep learning based image classification systems require large amount of training data and long training time. However, the availability of large annotated image dataset is usually limited and expensive to generate, which limits a vision system to adapt to new task efficiently. In this paper, a few-shot classification framework is proposed which can adapt one fruit ripeness classification system to classify new types of fruits using only a few training samples. The proposed framework adopts the meta-learning paradigm where a base network learns to extract meta-features and few-shot classification tasks from the base classes with abundant training samples and then apply the network to similar task on the novel classes using only a few support samples. Experimental results indicate that the proposed framework is able to achieve over 75% ripeness classification accuracy on various fruits using a little as five samples.

KW - Classification

KW - Deep learning

KW - Few-shot learning

KW - Fruit ripeness

UR - http://www.scopus.com/inward/record.url?scp=85125245558&partnerID=8YFLogxK

U2 - 10.1007/978-981-16-8129-5_109

DO - 10.1007/978-981-16-8129-5_109

M3 - 會議論文篇章

AN - SCOPUS:85125245558

SN - 9789811681288

T3 - Lecture Notes in Electrical Engineering

SP - 715

EP - 720

BT - Proceedings of the 11th International Conference on Robotics, Vision, Signal Processing and Power Applications - Enhancing Research and Innovation through the Fourth Industrial Revolution

A2 - Mahyuddin, Nor Muzlifah

A2 - Mat Noor, Nor Rizuan

A2 - Mat Sakim, Harsa Amylia

PB - Springer Science and Business Media Deutschland GmbH

T2 - 11th International Conference on Robotics, Vision, Signal Processing and Power Applications, RoViSP 2021

Y2 - 5 April 2021 through 6 April 2021

ER -

Ng HF, Lo JJ, Lin CY, Tan HK, Chuah JH, Leung KH. Fruit Ripeness Classification with Few-Shot Learning. In Mahyuddin NM, Mat Noor NR, Mat Sakim HA, editors, Proceedings of the 11th International Conference on Robotics, Vision, Signal Processing and Power Applications - Enhancing Research and Innovation through the Fourth Industrial Revolution. Springer Science and Business Media Deutschland GmbH. 2022. p. 715-720. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-981-16-8129-5_109

Fruit Ripeness Classification with Few-Shot Learning

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this