Object Detection: Custom Trained Models for Quality Monitoring of Fused Filament Fabrication Process

Abstract

Process reliability and quality output are critical indicators for the upscaling potential of a fabrication process on an industrial level. Fused filament fabrication (FFF) is a versatile additive manufacturing (AM) technology that provides viable and cost-effective solutions for prototyping applications and low-volume manufacturing of high-performance functional parts, yet is defect-prone due to the inherent aspect of parametrization. A systematic yet parametric workflow for quality inspection is therefore required. The work presented describes a versatile and reliable framework for automatic defect detection during the FFF process, enabled by artificial intelligence-based computer vision. Specifically, state-of-the-art deep learning models are developed for in-line inspection of individual thermoplastic strands’ surface morphology and weld quality, thus defining acceptable limits for FFF process parameter values. We examine the capabilities of an NVIDIA Jetson Nano, a low-power, high-performance computer with an integrated graphical processing unit (GPU). The developed deep learning models used in this analysis use a pre-trained model combined with manual configurations in order to efficiently identify the thermoplastic strands’ surface morphology. The proposed methodology aims to facilitate process parameter selection and the early identification of critical defects, toward an overall improvement in process reliability with reduced operator intervention.

The Influence of Thermoplastic Composite Recycling on the Additive Manufacturing Process and In-Use Phase as Candidate Materials for Wearable Devices Applications

“…To address this, we propose automation through image analysis and machine learning as a potential avenue for further research. By automating these processes, we can enhance objectivity and accuracy when assessing defects and optimizing printing parameters [ 35 ]. However, the purpose of the current study is to establish a straightforward workflow for material evaluation and the printing parameter optimization of the different recycling cycles of composite filaments based on material performance, as observed through stereoscope, as an efficient and cost-effective alternative to extended material testing and characterization methods.…”

A Real-Time Defect Detection Strategy for Additive Manufacturing Processes Based on Deep Learning and Machine Vision Technologies

“…Moreover, these methods have also been applied in the AM area for defect detection. For example, G. Bakas et al proposed a computer vision-based method for automatic defect detection in the fused deposition modelling (FDM) process [ 20 ]. Xu et al developed an improved one-stage model based on the You Only Look Once (YOLO) v4 to detect the print quality of the FDM process [ 21 ].…”

“…Moreover, these methods have also been applied in the AM area for defect detection. For example, G. Bakas et al proposed a computer vision-based method for automatic defect detection in the fused deposition modelling (FDM) process [20]. Xu et al.…”

Object detection model-based quality inspection using a deep CNN

“…2,3 The previous few years have seen a tremendous development in deep learning algorithms which have been used for quality control. The work of Bakas et al 4 presented a framework for automatic defect detection during the fused filament fabrication process using a deep learning architecture. The authors examined the capabilities of an NVIDIA Jetson Nano, 5 a low-power, highperformance computer with an integrated graphical processing unit (GPU).…”