pcb-defect-detection

An end-to-end deep learning system for automated PCB defect detection that combines computer vision with domain expertise. This project demonstrates the practical application of AI in industrial quality control, achieving 91.2% F1-score on multi-label defect classification.

Bare PCB defect detection using Image Subtraction technique, implemented with OpenCV in Python.

Modelling & Training for a AI-Driven PCB Fault Detection project.

PCB AI Inspector. PCB AI 缺陷检测. AI-powered PCB defect detection system based on YOLO11, supporting real-time inspection & static triggered detection.

YOLOv8x-HICAUps integrates a HorNet backbone, Convolutional Block Attention Module (CBAM), and attention-based up-sampling to improve feature extraction, feature fusion, and small-object detection accuracy while maintaining computational efficiency.

🛠️ Detect and classify PCB defects automatically using YOLO11 for efficient quality control and enhanced manufacturing processes.

An industrial-grade automated optical inspection (AOI) system for Printed Circuit Boards (PCBs). Features a computer vision pipeline for precise defect localization, a fine-tuned EfficientNetB0 model achieving 97.8% classification accuracy, and a full-stack Streamlit dashboard with real-time analytics, batch processing, and automated PDF reporting.

This repository contains the code and resources for a PCB defect detection project. The project uses YOLO and other comparative models to detect and classify PCB defects, along with improvements to the dataset for achieving better results.

Source-wise PCB defect detection benchmark with YOLOv11s, RT-DETR, Faster R-CNN, and a cleaned 6-class merged dataset.

PCB电路板缺陷检测系统，基于YOLOv8实现6种缺陷识别，mAP50=93.7%

A Flutter UI for AI-Driven PCB Fault Detection, with Rust+Ort used for the ML inference on edge.

Station de détection & diagnostic de défauts PCB — YOLOv11 (DsPCBSD+) + FastAPI + LLM + React

🟩 Bare PCB defect detection with YOLO26s — missing hole, mouse bite, open circuit, short, spur, spurious copper, pin hole. Test mAP50 0.928 / mAP50-95 0.617. Cleaned PKU-Market-PCB + DeepPCB merge, weights + Colab included.

🔬 PCB solder & assembly defect detection with YOLO26m — solder bridge, insufficient/excess solder, void, missing/shifted component, burn, damage. Test mAP50 0.895 / mAP50-95 0.657. Weights + Colab notebook included. AOI.

PCB Defect Detector designed to analyze and detect defects in PCBs. Leverages modern web technologies and tools to provide an intuitive interface for uploading, analyzing, and visualizing PCB defects. Also includes batch processing, dashboard analytics, and explainable AI insights. Next.js, Prisma, YOLO via Roboflow.

AI-Based PCB Defect Detection and Inspection System using YOLOv8 and Computer Vision

YOLOv8 六类 PCB 缺陷检测系统（鼠咬/毛刺/缺孔/短路/开路/多余铜），自研评估脚本实现 IoU 匹配与 P/R/F1 统计，检测结果经 UART 下发 STC89C52RC 单片机，LCD1602 上屏 + 红绿灯指示。测试集精确率 97.81%、召回率 99.22%、F1 98.51%。

AI-powered PCB defect detection and explanation system using YOLOv11 and TinyLlama.

Vision Transformer-based PCB defect detection system for automated inspection and quality control.

PCB defect detection and classification using a Custom CNN and MobileNetV2 transfer learning model.