Abstract:Visual surface defect inspection provides an important tool for product quality assessment in a wide range of industrial applications. In recent years, diverse convolutional neural network (CNN) models have been devel-oped for the high-accuracy image classification of natural objects. However, it remains a challenging task to develop deep-learning-based approaches for surface defect inspection of industrial products. The main challenge is that surface defects of industrial products occur in low probability, therefore it is impractical to build a large image dataset containing different types of defects to train/fine-tune deep CNN models to achieve satisfactory performance when applied to industrial inspection tasks. To overcome the above-mentioned problem, we build a light-weight defect classification model based on the pre-trained SqueezeNet architecture and present three effective techniques to achieve high-accuracy defect classification using very few collected defective training samples. First, we experimentally demonstrate that it is reasonable to freeze shallower convolutional layers and only fine-tune deeper layers in the last convolutional stage when there are not enough training samples. Second, we integrate batch-size independent Group Normalization (GN) into SqueezeNet to compute stable statistics based on a limited number of defective samples for training the classification model. Third, we add an auxiliary task to perform the similarity comparison of two defective samples during the training phase, significantly increasing the size of the training data and generating more distinctive defect-specific representations. The experimental results on the NEU-CLS dataset and USB-FS dataset verify the effectiveness of the proposed method. Our method can obtain accurate defect classification results when very few defective training samples are provided, i.e., about 97.69% and 82.92% average classification accuracy on the NEU-CLS and USB-FS datasets, respectively, under the 5-shot setting (5 images per defect category).

In-situ Quality Inspection System of Injection Parts Based on Transfer Learning

A Novel Method of Fault Diagnosis for Injection Molding Systems Based on Improved VGG16 and Machine Vision

Application of Mobile Injection Molding Pump Defect Detection System Based on Deep Learning

In-line Inspection Solution for Codes on Complex Backgrounds for the Plastic Container Industry

Smart defect detection using transfer learning in injection molding: a comparative exploration study of deep learning architectures

In Situ Quality Inspection with Layer-Wise Visual Images Based on Deep Transfer Learning During Selective Laser Melting

A novel DL-BiGRU multi-feature fusion and deep transfer learning based modeling approach for quality prediction of injection molded products using small-sample datasets

Small-scale defect detection in industrial environment based on lightweight deep learning network

Defect detection of injection molding products on small datasets using transfer learning

Multi-quality prediction of injection molding parts using a hybrid machine learning model

An Effective Industrial Defect Classification Method under the Few-Shot Setting Via Two-Stream Training

Deep Learning-Based Predictive Control of Injection Velocity in Injection Molding Machines

Online Assembly Inspection Integrating Lightweight Hybrid Neural Network With Positioning Box Matching

Industry 4.0 In-Line AI Quality Control of Plastic Injection Molded Parts

A Data-Centric Approach to Design and Analysis of a Surface-Inspection System Based on Deep Learning in the Plastic Injection Molding Industry

Induced network-based transfer learning in injection molding for process modelling and optimization with artificial neural networks

Artificial neural network-based online defect detection system with in-mold temperature and pressure sensors for high precision injection molding

Transfer learning on convolutional activation feature as applied to a building quality assessment robot

An Intelligent Foreign Substance Inspection Method for Injection Based on Machine Vision

A Lightweight Deep Network for Defect Detection of Insert Molding Based on X-ray Imaging

Real-Time Quality Inspection Based on Transfer Learning and Feature Clustering for Wave Soldering