INTRODUCE DEEP LEARNING AI TO YOUR ADDITIVE MANUFACTURING R&D AND QC Hybridization of DL and traditional imaging creates advantage. SPONSORED CONTENT For more information on deep learning image analysis solutions for AM, contact MIPAR Software at support@mipar.us / 614.407.4510 / www.mipar.us. In today’s fast-moving environment, reducing project turnaround time, accelerating research and development, and meeting productivity targets, all while reducing operating cost can be challenging without a smart automated approach. To avoid product recalls, meet increased customer demands, and continually innovate, a thorough investigation of materials’ microstructure is key. Defect, inclusion, and grain size analysis are only a few approaches that can give meaningful insights into the quality of products. Modern research and quality control have the unique challenge of working with real world, imperfect micrographs that require a flexible tool suite. MIPAR Image Analysis combines customized algorithms and powerful deep learning systems to produce technology able to perform sophisticated structure investigation. Whether of titanium, copper, steel, aluminum, or ceramics, MIPAR’s software allows for automated micrograph analysis that streamlines data analytics, improves data quality, and offers new layers of information. Automation re- duces operator error and improves professional productivity. A primary challenge in modern R&D and QC processes is that the manufacturing environment is increasingly driven by stringent efficiency requirements in the name of productivity. Guaranteeing product quality often runs contrary to pushing the bottom line, meaning defect and contaminant analysismust be carried out quickly as well as effectively. Automation and digital integration are central to the push for greater productivity in manufacturing environments. The concept of automating production typically brings to mind the robotic arms of assembly lines, but manufacturers are just as interested in software solutions that accelerate critical processes throughout the manufacturing pipeline. Deep learning is one such solution. Did you know that more and more companies are using deep learning to double check the material quality provided by suppliers? Be ahead of your customers by introducing these capabilities in-house. Improve your customer satisfaction, avoid rework, while reducing the operational cost. What is Deep Learning? The aimof deep learning AI (artificial intelligence) is to teach software to adapt to your own micrographs. It works in the presence of varying contrasts and feature texture, as well as sample preparation artifacts. As little as four images can be used to train an application specific solution. This can be done with minimum training and no programming expertise. This technology has had a profound role in developing the latest micrograph analysis solutions in the additive manufacturing (AM) space. Not only does AM imagery suffer from the usual challenges of varying contrast/ lighting, sample prep noise, etc., it offers especially complex microstructural features, often with extremely poor contrast due to the rapid cooling rate and high deformation processes involved in part fabrication. While humans have proven adept at identifying these features by eye, traditional automated software has strug- gled handsomely. Breakthrough Solutions MIPAR’s unique hybridization between deep learning and traditional image processing has allowed for rapid custom development of breakthrough solutions for automated AM materials analysis and inspection. Examples include melt pool quantification, defect classification, layer thickness profiling, ultrafine phase measurement, complex grain sizing, virtual powder precursor sieve analysis, and part porositymapping, just to name a few. Fully automated particle detection and satellite classification. Enables size and shape distribution per particle class. Advanced automation of complex twinned grain size analysis in aged additively manufactured component.
RkJQdWJsaXNoZXIy MTYyMzk3NQ==