High Tech – Icuro

Advancing Silicon Innovation With Artificial Intelligence

Silicon Everywhere

Silicon is the heart in every aspect of daily life of humans. The emerging mega market trends is seeking a symbiotic ecosystem relationship from silicon to chip and integrated software to innovate products and services. Artificial intelligence (AI) is the key accelerator to realize this market need.

ICURO has made significant investments in developing and deploying AI platforms with applications ranging from design-for-silicon to silicon production process. The silicon functional and technical AI practitioners at ICURO have deployed innovative and integrated platforms and modular systems to transform and realize key outcomes across each and every stage of the silicon value chain.

Adaptive AI Platform For Consumer Products

The adaptive intelligence platform for consumer products ecosystem empowers retailers, manufacturers, and supply chain logistics to make insightful decisions at different stages from product innovation to shop floor to consumption. The platform leverages data based on commerce transactions at outlets. Each outlet needs specific personalized offers to consumers as consumers are demanding owing to the influence of social-economic dynamic fluctuations. Precise suggested orders will enable long term relationships through intrusion free, hassle free and quality products and usage. This is made possible by augmented intelligence from product innovation to purchase at every stage for every stakeholder. This improves production quality, purchase patterns, cognitive decision making and reduced recalls. This experience is delivered by an intelligent workflow of the consumer lifecycle from product innovation to shopping.

Intelligent Visual Wizard

The no-code AI vision platform with a user-friendly visual builder provides precision analysis and measurement of desirable features in silicon. It is a self-learning AI system embedded in the manufacturing process to deliver extreme high precision and accuracy by eliminating manual steps at every stage of the silicon manufacturing innovation process.
The platform empowers the lab technicians, quality engineers, and production managers with a rapid inline visual analysis and active feedback system to advance yield improvements. Some of the features include the following

Create a new project and choose the semiconductor production line association
Platform recommends prebuilt models related to yield improvement, wafer inspection, production optimization
Load images with dynamic annotation with auto model creation
Ground truth update for the image analysis
Summary report of the inference for the entire image analysis

Consumable Multimodal Store

The adaptive AI technology platform integrates with silicon design and analysis toolkit, semiconductor manufacturing equipment, metrology, and visual inspection and measurement subsystems. The adaptive AI techniques remove the extensive labeling of thousands of images or unstructured data that is typically required for a supervised learning pipeline. The production-grade and scalable AI models are best suited for the design lab and shop floor technicians to augment their analysis and reduce manual efforts.
The ready-to-deploy extensible multimodal addresses the following

Yield in semiconductor manufacturing is key and it is a time-consuming process. AI predicts failure in new designs, identifies patterns in component failures, and recommends optimal layouts to improve yield
Predictive analytics for anomaly detection in operations coupled with natural language processing based on knowledge graph with industrial IoT, and equipment data
Location and tracking of items for inventory planning and demand forecasting

End-to-End Pipeline Automation

The intelligent workflow automates every stage from silicon design to semiconductor manufacturing process. It empowers the workforce with digital tools and real time knowledge to adapt, collaborate, and excel in fast-changing conditions. In the wafer manufacturing process, the entire pipeline consisting of contour extraction, template matching, and alignment eliminates the manual steps for characterizing semiconductor geometry and features by applying advanced deep learning to automate measurements of wafer microscopy images. The adapters developed for image capture from atomic force microscope, scanning electron microscope, and transmission electron microscope camera manufacturers. The end-to-end AI pipeline offers both real-time and batch processing for wafer images and measurement of key specifications.

Business Outcomes Driven

Production yield is the key critical success factor in controlling wafer processing costs. It affects output per unit of input resources, waste generation, environmental and financial sustainability, time to market, and costs. Yield improvements provides significant competitive advantages, which are critical in unpredictable economic environments. Production yield improvement can provide cost competitiveness through optimum resource utilization along with waste reduction and achieving sustainable operations. The AI platform has demonstrated consistent business outcomes in the following three critical success factors.

Strategic Partners

Associate Member of SEMI, the global industry association representing semiconductor and electronics manufacturing and design supply chain. ICURO contributes to SEMI Smart Initiatives that advances technology around semiconductors, electronics and imaging systems, the Internet of Things, MEMS, sensors, devices, displays, and other digital technologies used in the electronics industry.
As an Associate Member of SEMI, ICURO is an active innovation partner to the following SEMI Smart Initiatives:

Smart Data. This initiative aims to improve efficiency in the semiconductor design and manufacturing supply chain through new data analytics, artificial intelligence (AI) and machine learning (ML). In addition to efficiency improvements, activities aim to reduce costs, validate semiconductor chips and products, speed process development and to problem-solve root causes.
Smart Manufacturing. This initiative focuses on challenges and opportunities integrating production and sensor data, analytics, artificial intelligence, automated systems and more with traditional manufacturing technologies.
Smart MedTech. Nano-Bio Materials Consortium of SEMI brings critical information and project funding to the electronics industry; the SEMI-FlexTech partnership focuses on the rise of flexible, printed and hybrid electronics to better fit the contours and movement of the human form, in part for healthcare applications.
Smart Mobility. This initiative is focused primarily on the automotive and autonomous vehicle supply chain. The Global Automotive Advisory Council of SEMI engages stakeholders to address common challenges, priorities, solutions, and opportunities

Strategic Partner of AMD MIVisionX, the comprehensive toolkit consisting of computer vision and machine intelligence libraries, utilities, and applications bundled into a single toolkit. AMD MIVisionX delivers highly optimized conformant open-source implementation of the Khronos OpenVX™ and OpenVX™ Extensions along with Convolution Neural Net Model Compiler & Optimizer supporting ONNX, and Khronos NNEF™ exchange formats. The toolkit allows for rapid prototyping and deployment of optimized computer vision and machine learning inference workloads on a wide range of computer hardware, including small embedded x86 CPUs, APUs, discrete GPUs, and heterogeneous servers.

As a strategic partner, ICURO has been applying AMD Radeon Open Compute (ROCm) and MIVisionX to deliver advanced accelerators to effectively use OpenVX for feature detection and OpenCL for GPU acceleration in Simultaneous Localization and Mapping (SLAM) applications. SLAM is a computational problem for constructing a 3D map of an unknown area while still knowing where the machine or robot exists in the map and this accelerator developed by ICURO is used for a wide range of applications such as 3D perception for machines, robots, and drones.