Abstract:Ceramic substrates serve as the foundational material for numerous electronic devices, and their surface quality directly affects performance and longevity. Therefore, surface defect detection on ceramic substrates is an indispensable task during the manufacturing process. Ceramic substrate surface defects exhibit low contrast, along with intraclass difference and interclass similarity, posing substantial challenges for automated visual inspection. Existing deep learning-based methods utilize carefully crafted backbone networks and feature pyramid structures that perceive multilevel information to deal with the above challenges. However, these methods still suffer from insufficient extraction of discriminative features and low effectiveness in feature fusion, thereby hindering the detection performance. To address these issues, we propose the profound feature exploration and interaction network (PFEI-Net). Initially, we propose the discriminative feature mining backbone (DFM-backbone) to progressively explore the effective discriminative features. In the DFM-backbone, the global information grated perception module is devised to construct long-range feature dependencies and aggregate rich features, whereas the contextual semantic flexible extraction module is designed to meticulously perceive contextual semantic information in a targeted manner, enhancing the comprehensive representation of the discriminative features. Then, we propose the multipath semantic interaction guided-feature pyramid network (MSIG-FPN) to facilitate the fusion and interaction of the valuable information. In the MSIG-FPN, the hierarchical refocus–reaggregation module is constructed to refocus on task-relevant features and provide reliable guidance for the network, whereas the cross-stage semantic deep fusion module is designed to integrate multi-level contextual information deeply across stages and is employed to construct the semantic complementary paths, thereby preventing the loss of crucial features. Finally, built upon the backbone of DFM-backbone and MSIG-FPN, we establish the PFEI-Net, achieving accurate detection of surface defects on ceramic substrates. Experimental results highlight that the PFEI-Net achieves excellent performance with a remarkable 89.3 % mean average precision (mAP) and an inference time of 26.1 ms, which offers a promising and viable solution for automated surface defect detection of ceramic substrate.

Micro-defect Detection Based on Multi-scale Feature Backtracking Structure

A lightweight and efficient model for surface tiny defect detection

Automated Detection of Defects with Low Semantic Information in X-ray Images Based on Deep Learning

A Small-Sized Object Detection Oriented Multi-Scale Feature Fusion Approach With Application to Defect Detection

PGA-Net: Pyramid Feature Fusion and Global Context Attention Network for Automated Surface Defect Detection

Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale Detector

Hierarchical Multi-Scale Network for Cross-Scale Visual Defect Detection

Enhanced Darknet53 Combine MLFPN Based Real-Time Defect Detection in Steel Surface

PFEI-Net: A Profound Feature Exploration and Interaction Network for Ceramic Substrate Surface Defect Detection

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

A Method of Defect Detection for Focal Hard Samples PCB Based on Extended FPN Model

Joining Spatial Deformable Convolution and a Dense Feature Pyramid for Surface Defect Detection

A Multiple Feature-Maps Interaction Pyramid Network for Defect Detection of Steel Surface

PDD-Net: A Precise Defect Detection Network Based on Point Set Representation

Detection of PCB Surface Defects With Improved Faster-RCNN and Feature Pyramid Network

An enhanced method for surface defect detection in workpieces based on improved MobileNetV2‐SSD

An End-to-End Steel Surface Defect Detection Approach via Fusing Multiple Hierarchical Features

Surface defect detection of industrial components based on vision

TDD-net: a Tiny Defect Detection Network for Printed Circuit Boards

ES-Net: Efficient Scale-aware Network for Tiny Defect Detection