Improving crop image recognition performance using pseudolabels

In crop image recognition, when faced with a large quantity of unlabeled data, the traditional manual labeling method requires a large amount of human and material resources. To solve this problem, this study proposes an image recognition method based on a pseudolabeling technique. First, the data a...

Full description

Saved in:
Bibliographic Details
Main Authors: Pengfei Deng, Zhaohui Jiang, Huimin Ma, Yuan Rao, Wu Zhang
Format: Article
Language:English
Published: Elsevier 2025-03-01
Series:Information Processing in Agriculture
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2214317324000015
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849708325885706240
author Pengfei Deng
Zhaohui Jiang
Huimin Ma
Yuan Rao
Wu Zhang
author_facet Pengfei Deng
Zhaohui Jiang
Huimin Ma
Yuan Rao
Wu Zhang
author_sort Pengfei Deng
collection DOAJ
description In crop image recognition, when faced with a large quantity of unlabeled data, the traditional manual labeling method requires a large amount of human and material resources. To solve this problem, this study proposes an image recognition method based on a pseudolabeling technique. First, the data are divided into labeled and unlabeled data. The initial network model is trained on labeled data. Then, pseudolabeling of the unlabeled data is predicted, and only the data that satisfy the confidence threshold are regarded as valid pseudolabeling. To convert the unlabeled data into supervised training data, the two types of data are mixed. The training is terminated when the number of remaining unlabeled data satisfies the end condition and when the fivefold cross-validation method is used to evaluate model performance. Compared with the traditional semisupervised method, the experimental method is simpler and more applicable. Experiments were conducted on rice growth stage recognition and crop weed seedling recognition tasks. The results showed that the proposed method achieved 99.17% accuracy in rice growth stage recognition and a high AUC value of 99.93% in crop weed seedling recognition, which demonstrated excellent performance. Compared with the traditional model, this method not only improves in accuracy but also has better stability and wider applicability and is expected to provide an efficient, accurate and scalable solution for crop image recognition.
format Article
id doaj-art-c8ef193c84b74b0582d76dd8afb660fc
institution DOAJ
issn 2214-3173
language English
publishDate 2025-03-01
publisher Elsevier
record_format Article
series Information Processing in Agriculture
spelling doaj-art-c8ef193c84b74b0582d76dd8afb660fc2025-08-20T03:15:42ZengElsevierInformation Processing in Agriculture2214-31732025-03-01121172610.1016/j.inpa.2024.02.001Improving crop image recognition performance using pseudolabelsPengfei Deng0Zhaohui Jiang1Huimin Ma2Yuan Rao3Wu Zhang4School of Information & Computer, Anhui Agricultural University, Hefei 230036, ChinaCorresponding author.; School of Information & Computer, Anhui Agricultural University, Hefei 230036, ChinaSchool of Information & Computer, Anhui Agricultural University, Hefei 230036, ChinaSchool of Information & Computer, Anhui Agricultural University, Hefei 230036, ChinaSchool of Information & Computer, Anhui Agricultural University, Hefei 230036, ChinaIn crop image recognition, when faced with a large quantity of unlabeled data, the traditional manual labeling method requires a large amount of human and material resources. To solve this problem, this study proposes an image recognition method based on a pseudolabeling technique. First, the data are divided into labeled and unlabeled data. The initial network model is trained on labeled data. Then, pseudolabeling of the unlabeled data is predicted, and only the data that satisfy the confidence threshold are regarded as valid pseudolabeling. To convert the unlabeled data into supervised training data, the two types of data are mixed. The training is terminated when the number of remaining unlabeled data satisfies the end condition and when the fivefold cross-validation method is used to evaluate model performance. Compared with the traditional semisupervised method, the experimental method is simpler and more applicable. Experiments were conducted on rice growth stage recognition and crop weed seedling recognition tasks. The results showed that the proposed method achieved 99.17% accuracy in rice growth stage recognition and a high AUC value of 99.93% in crop weed seedling recognition, which demonstrated excellent performance. Compared with the traditional model, this method not only improves in accuracy but also has better stability and wider applicability and is expected to provide an efficient, accurate and scalable solution for crop image recognition.http://www.sciencedirect.com/science/article/pii/S2214317324000015CropsImage recognitionDeep learningTransfer learningPseudolabel
spellingShingle Pengfei Deng
Zhaohui Jiang
Huimin Ma
Yuan Rao
Wu Zhang
Improving crop image recognition performance using pseudolabels
Information Processing in Agriculture
Crops
Image recognition
Deep learning
Transfer learning
Pseudolabel
title Improving crop image recognition performance using pseudolabels
title_full Improving crop image recognition performance using pseudolabels
title_fullStr Improving crop image recognition performance using pseudolabels
title_full_unstemmed Improving crop image recognition performance using pseudolabels
title_short Improving crop image recognition performance using pseudolabels
title_sort improving crop image recognition performance using pseudolabels
topic Crops
Image recognition
Deep learning
Transfer learning
Pseudolabel
url http://www.sciencedirect.com/science/article/pii/S2214317324000015
work_keys_str_mv AT pengfeideng improvingcropimagerecognitionperformanceusingpseudolabels
AT zhaohuijiang improvingcropimagerecognitionperformanceusingpseudolabels
AT huiminma improvingcropimagerecognitionperformanceusingpseudolabels
AT yuanrao improvingcropimagerecognitionperformanceusingpseudolabels
AT wuzhang improvingcropimagerecognitionperformanceusingpseudolabels