Search for: All records

Abstract The performance of conventional image processing techniques is highly dependent on many parameters like image quality, light source, background surface texture, optimal threshold value and particle morphology. However, during intermediate stages of manufacturing processes (such as continuous deposition, coating, mixing, and transfer), complex backgrounds can arise from heterogeneous particle-substrate (HPS) systems. In such HPS environments, particles become integrated with substrates or suspended in liquid carriers or etching media, making them challenging to identify using traditional particle analysis tools and techniques. In response to this challenge, a deep learning object detection algorithm (YOLO) has been put into practical use. Initially, an HPS (heterogeneous particle-substrate) system was created using a wet-deposition particle transfer process that involved the immersion of poly-disperse particles on to a cylindrical substrate. By manipulating the capillary number in the wet-deposition process, four distinct HPS morphologies were captured, each characterized by variations in image heterogeneity. These morphologies were subsequently subjected to detailed analysis with neural network-based AI algorithm. The proposed artificial intelligence tool has demonstrated an impressive ability to identify and analyze poly-dispersed particles within HPS morphologies, achieving an accuracy rate of over 97%. We can evaluate the quality of sorting by calculating the particle size distribution using the proposed method and find the ideal process parameters for the particle transfer process. The results of this study, outlined in this paper, underscore the potential of deep learning as a particle analysis tool for in-situ applications, even in environments with heterogeneous backgrounds. This developed tool holds promise for various manufacturing processes, including semiconductor industries, high-density powder-based 3D printing, powder metallurgy, refractory coatings in harsh environments, and particle sorting, among others. 

Abstract We propose a novel framework that combines state-of-the-art deep learning approaches with pre- and post-processing algorithms for particle detection in complex/heterogeneous backgrounds common in the manufacturing domain. Traditional methods, like size analyzers and those based on dilution, image processing, or deep learning, typically excel with homogeneous backgrounds. Yet, they often fall short in accurately detecting particles against the intricate and varied backgrounds characteristic of heterogeneous particle–substrate (HPS) interfaces in manufacturing. To address this, we've developed a flexible framework designed to detect particles in diverse environments and input types. Our modular framework hinges on model selection and AI-guided particle detection as its core, with preprocessing and postprocessing as integral components, creating a four-step process. This system is versatile, allowing for various preprocessing, AI model selections, and post-processing strategies. We demonstrate this with an entrainment-based particle delivery method, transferring various particles onto substrates that mimic the HPS interface. By altering particle and substrate properties (e.g., material type, size, roughness, shape) and process parameters (e.g., capillary number) during particle entrainment, we capture images under different ambient lighting conditions, introducing a range of HPS background complexities. In the preprocessing phase, we apply image enhancement and sharpening techniques to improve detection accuracy. Specifically, image enhancement adjusts the dynamic range and histogram, while sharpening increases contrast by combining the high pass filter output with the base image. We introduce an image classifier model (based on the type of heterogeneity), employing Transfer Learning with MobileNet as a Model Selector, to identify the most appropriate AI model (i.e., YOLO model) for analyzing each specific image, thereby enhancing detection accuracy across particle–substrate variations. Following image classification based on heterogeneity, the relevant YOLO model is employed for particle identification, with a distinct YOLO model generated for each heterogeneity type, improving overall classification performance. In the post-processing phase, domain knowledge is used to minimize false positives. Our analysis indicates that the AI-guided framework maintains consistent precision and recall across various HPS conditions, with the harmonic mean of these metrics comparable to those of individual AI model outcomes. This tool shows potential for advancing in-situ process monitoring across multiple manufacturing operations, including high-density powder-based 3D printing, powder metallurgy, extreme environment coatings, particle categorization, and semiconductor manufacturing.