Automated vision-based inspection of mould and part quality in soft tooling injection moulding using imaging and deep learning

Article Computer Science, Software Engineering

A robust multiclass 3D object recognition based on modern YOLO deep learning algorithms

Mariam L. Francies et al.

Summary: This study compares the performance and characteristics of YOLOv3, YOLOv4, and YOLOv5 in multiclass 3D object detection and recognition. Experimental results show that YOLOv3 has the highest recognition accuracy, while the modified YOLOv5 has the shortest processing time.

CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE (2022)

添加到收藏夹

Article Engineering, Industrial

Evolution and future of manufacturing systems

Hoda ElMaraghy et al.

Summary: The article discusses the impact of changes in products, technology, business strategies, and production paradigms on the evolution of manufacturing systems, as well as the trend of adopting human-centric decision making in meshed collaboration with intelligent systems. It also reviews the implications and preparedness for the shift towards more responsive, intelligent adaptive systems, and presents research and industrial use cases. Furthermore, the article articulates a vision for the new Adaptive Cognitive Manufacturing System (ACMS) paradigm, highlighting its characteristics, drivers, and enablers in the digital and cognitive transformation.

CIRP ANNALS-MANUFACTURING TECHNOLOGY (2021)

添加到收藏夹

Article Engineering, Industrial

Augmented semantic segmentation for the digitization of grinding tools based on deep learning

Petra Wiederkehr et al.

Summary: The paragraph introduces the use of grinding simulations to analyze various process characteristics. In order to accurately analyze the tool and individual grains, the grinding tools need to be digitized. A deep learning-based methodology is developed to efficiently achieve high segmentation accuracy of the grain boundaries, and a data augmentation approach is investigated to limit the necessary data for learning. The model transferability is quantified by analyzing different states of tool wear.

CIRP ANNALS-MANUFACTURING TECHNOLOGY (2021)

添加到收藏夹

Article Computer Science, Theory & Methods