Research on Surface Defects Classification for PET Preform by Fusing Multi-Scale Features
Open Access
- 16 July 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 134312-134324
Abstract
Product quality control is an important component of PET preform production process. In order to improve the accuracy and reliability of PET preform detection, reduce the cost of product quality defects, and improve the overall level of product quality, this paper proposes a surface defect detection method for PET preform based on multi-scale feature fusion. Multi-scale features fusion combines features from different scales to produce more accurate and robust feature representations, which improve the accuracy, stability and adaptability of PET preform detection model. This feature processing model constructs a three-branch network, generating feature maps of different sizes. Then, the low-level feature, the middle-level feature and the high-level feature are fused to obtain a PET preform feature map based on multi-scale feature fusion by using a maximum value fusion strategy. The original image and feature map of the PET preform are respectively input into the deep convolutional neural network for feature extraction and fusion. And finally the fusion features are input into the ECOC-SVM to achieve surface defect classification of the PET preform. The machine vision platform is designed for collecting and detecting PET preform data. By importing the model designed in this paper, surface defect recognition and defect category judgment of PET preform can be achieved. Through experimental data analysis, the higher testing detection accuracy of this method is 98.8%, with an average testing accuracy of 97.1%.Keywords
Funding Information
- program for scientific research start-up funds of Guangdong Ocean University (060302062311, 060302062315)
- Ministry of Education’s Industry University Cooperation Collaborative Education Project in batches in 2023 (230802784214645, 230803817223502, 230821384507208)
This publication has 36 references indexed in Scilit:
- Fabric defect detection and classification via deep learning-based improved Mask RCNNSignal, Image and Video Processing, 2023
- D3SENet: A hybrid deep feature extraction network for Covid-19 classification using chest X-ray imagesBiomedical Signal Processing and Control, 2023
- A fast and time-efficient glitch classification method: A deep learning-based visual feature extractor for machine learning algorithmsAstronomy and Computing, 2022
- An early warning model for the stuck-in medical drilling process based on the artificial fish swarm algorithm and SVMDistributed and Parallel Databases, 2022
- Tuning of data augmentation hyperparameters in deep learning to building construction image classification with small datasetsInternational Journal of Machine Learning and Cybernetics, 2022
- Tuning hyperparameters of machine learning algorithms and deep neural networks using metaheuristics: A bioinformatics study on biomedical and biological casesComputational Biology and Chemistry, 2022
- Efficacy of Transfer Learning-based ResNet models in Chest X-ray image classification for detecting COVID-19 PneumoniaChemometrics and Intelligent Laboratory Systems, 2022
- AI classification of wafer map defect patterns by using dual-channel convolutional neural networkEngineering Failure Analysis, 2021
- Classification of oils by ECOC based multi-class SVM using spectral analysis of acoustic signalsApplied Acoustics, 2021
- Surface defect classification and detection on extruded aluminum profiles using convolutional neural networksInternational Journal of Material Forming, 2019