Скачать или смотреть J015: Deep learning methods for object detection in smart manufacturing: A survey

Ahmad H.*, Rahimi A. (2022, Jul.) Deep Learning Methods for Object Detection in Smart Manufacturing: A Survey, Journal of Manufacturing Systems, from Elsevier BV, Volume 64, Issue N/A, pp. 181-196, doi: 10.1016/j.jmsy.2022.06.011

This academic review examines the evolution and implementation of deep learning-based object detection within the realm of smart manufacturing. The authors highlight how Convolutional Neural Networks (CNNs) have superseded traditional machine learning by effectively processing massive volumes of industrial data to improve quality control and automated inspection. By analyzing various backbone architectures—such as ResNet, VGG, and MobileNet—the text illustrates how these models achieve high precision in identifying defects and monitoring assembly line tasks. Furthermore, the survey categorizes major region selection methods, including the R-CNN family and YOLO, while comparing their trade-offs between speed and accuracy. Ultimately, the source serves as a comprehensive guide for utilizing artificial intelligence to reduce manufacturing costs and enhance operational productivity.