English title

Here’s the translation for your abstract title and keywords:

Title:
Application and Evaluation of Deep Learning Models for Defect Detection in Printed Circuit Boards

Abstract:
In this paper, the effectiveness of deep learning models for the automatic detection of defects in printed circuit boards is investigated. Defects in printed circuit boards can lead to serious disruptions in the performance of electronic devices, and accurate and rapid identification of these defects is essential for ensuring production quality. To this end, the Inception-v3, VGG16, ResNet18, ResNet50, ResNet101, and YOLOv5 models were selected as representatives of various deep neural network architectures, and trained and evaluated on datasets of images containing PCB defects. These models, due to their capabilities in extracting complex and deep features from images, facilitate precise detection of both surface and structural defects. The results from evaluating these models indicate that deep learning can accurately and automatically identify defects in printed circuit boards, thereby improving the inspection process. This research emphasizes the importance of utilizing advanced deep learning models in the electronics industry and demonstrates that the choice of appropriate architecture significantly impacts the accuracy and speed of defect detection.