AEWIN cooperates with UNISEM’s IoT Division to provide the UniTraffic, an AI (Deep Learning)-based traffic management solution, to help in solving traffic problems caused by complex traffic environment and loose adherence to traffic regulations. Traffic is a major issue affecting metropolitan cities and UniTraffic can assist in managing traffic to improve the operations across the city.

UniTraffic’s core technologies are Deep Learning-based image processing technology and multi-object detection that allows counting and classifying on-road objects such as pedestrians, automobiles, motorcycles, buses and trucks. It has 99% accuracy for object recognition in real-time and due to multiple object tracking can identify traffic violations simultaneously.

UniTraffic’s Video Analytics System, VAS, capable of detecting objects from video feeds and classify the objects according to the features of objects through Deep Learning. LPDR, License Plate Detection and Recognition, detects the license plate number of vehicles with violations from the video input. TMS, Transportation Management System, is the backend web-based system to analyze, monitor, and aggregate traffic information by collecting the data from the VAS & LPDR. TMS can provide users with the data on traffic analysis and police enforcement result. UniTraffic solution is useful for city planners by providing statistics and traffic flow patterns to plan future infrastructural development and upgrades. Likewise for city administration and police departments, UniTraffic can minimize time used for administering tickets and reduce personnel costs.

UniTraffic requires a high-performance system to process multiples of live video inputs.  AEWIN SCB-1932 is the ideal platform to host this application. It is powered by the latest Intel 3rd Generation Xeon-SP processors, support up to 2TB of memory, and 1600W redundant power supplies. The design is a refinement of our successful and widely deployed SCB-1921 network computing system. Video input processing is provided by up to 4x NVIDIA T4 GPUs. SCB-1932 has on-board BMC for ease of management and AEWIN Secure Boot technology for safe-guarding on-board firmware. Through AEWIN’s network computing lineage, SCB-1932 has outstanding network connectivity options through AEWIN’s full line of network expansion modules that ranges from 1Gbps all the up to 100Gbps.

Through personalized collaboration, it enables us to fine tune performance of the servers and GPU for application specific workloads. This is the difference between going with a general purpose server from other manufacturers versus AEWIN. See how AEWIN and UNISEM together can help you leap into the future of city traffic management.

SCB-1932 Dual 3rd Gen Intel Xeon SP High Performance Edge Computing system 8x AEWIN Expansion Module bays, supporting Network Expansion Modules, GPU, U.2 Convertors ● Support 3rd Gen Intel Xeon Scalable Processor(Ice Lake-SP) ● Support Max 2TB DDR4 3200MHz & Optane DC Persistent Memory ● Max support 8x PCIe Gen4 x8 slots for Network Expansion Module ● Support BMC/IPMI& hardware One-Key Bypass function

這篇文章 AEWIN的UniTraffic解決方案 最早出現於 AEWIN。

To prevent various accidents during the manufacturing process, AEWIN cooperates with UNISEM’s IoT Division to provide UniSafety, an AI-based computer vision solution. This is a Smart Factory solution that uses Deep Learning-based computer vision technology to provide a safe work environment, prevent potential accident, and reduce serious injuries.

UniSafety is based on video analytics technology enhanced with deep learning that allows detection of various dangerous situations, such as fire, falling materials, abnormal process, crush injury danger, non-wearing of safety equipment, intrusion, etc., that can occur during the production process. In the case of an accident or unusual situation, the system informs the safety supervisors via web-dashboard notification, text message, and loudspeaker. It also offers motion detection for detecting unauthorized access or intrusions. This can allow a more open working space without using barriers to keep out intruders.

UniSafety analyzes real-time video feeds to surpass the limitations of visual supervision on the shop floor, so all the situations on the factory site can be detected automatically. UniSafety automates monitoring and operates 24/7 to allow the supervisory staff to maintain focus and reduce incident response time. This in term helps to protect the factory floor workers, reduce damages to equipment, and minimize potential down times. UniSafety provides hourly/daily conditional statistics for generating detailed safety management report. Supervisors and managers can access and monitor through the web-based dashboard from PC or smart phone browser to provide flexibility and ability to audit the factory floor from anywhere. This solution has uses outside of the factory, such as construction sites, shipyards and warehouses.

AEWIN BIS-3101 is a light and powerful accelerated edge platform optimized for UniSafety. It is outfitted with Intel 9^th Gen Coffee Lake 8 core CPU, dual-width FHFL NVIDIA GPU accelerators, 32GB memory, and two Ethernet for redundant network access. Through repeated testing and benchmarking, it enabled us to fine tune the performance of GPU accelerated BIS-3101 for application specific workloads. See how AEWIN and UNISEM together can help you increase the operational safety of your workspace.

BIS-3101 Coffee Lake 8 core Light Edge Computing system 32G memory, two 2.5 or 3.5 SSD/HDD, on board M.2 slot, support dual width FHFL GPU. ●Support Intel 8/9th Core i7/i5/i3 (Coffee Lake-S) series processor, LGA 1151 ●2x channel DDR4 2400MHz un-buffered / Non-ECC SODIMM, max support 32GB ●1x dual width PCIe Gen 3 x16 slot, supporting RTX-3070

這篇文章 AEWIN的UniSafety解決方案 最早出現於 AEWIN。

AEWIN cooperates with D8AI to provide NLP Document Recognizer solution to transform unstructured information to useful data. It can reduce the operating cost, enhance the service, and accelerated document process time. Enterprise, hospital, or even government can improve the data processing quality with this solution.

NLP Document Recognizer includes a proprietary AI assisted data extraction technology which can transform multiple optical and text data type simultaneously. These formats include Chinese, English, barcode, QR code, and other digital data. It can help organizations speed up digitization of paper documents for processing and storage in the database. It utilizes deep learning and computer vision in unison to find structure in data, which can be used for predictive analytics. Additional data types and languages can be integrated with customization to fit any requirement.

NLP Document Recognizer processes multiple format image through proprietary D8AI CT-SSD and CTC-CNN algorithms for text detection and text recognition for model training. The model can be trained with data from specific domain knowledge, experiences, Q&A, regulations, and human feedbacks for continuous model improvement.

NLP Document Recognizer application needs the reliability of workstation-class computing system, such as the BIS-3101. It is powered by Intel 8th /9th Coffee Lake 8 core CPU, 32GB of memory, and support NVIDIA T4 or dual width FHFL GPU over 200W. For organization with large databases, real-time inferencing, massive batch processing, and high complexity data models, the SCB-1932 is the answer. With dual Intel 3rd Xeon Scalable Processors, up to 2TB of memory, and 2x dual width datacenter focused NVIDIA A30 GPU, this is a powerhouse system that can accommodate any organizational requirements.

AEWIN is specialized in Edge and network computing platforms. Our edge computing platforms have been optimized for accelerated computing with NVIDIA GPU hardware. With optimized BIOS and careful component selection, we are able to achieve best results to meet customer requirements for NLP Document Recognizer for any scale of deployment.

BIS-3101 Coffee Lake 8 core Light Edge Computing system 32G memory, two 2.5 or 3.5 SSD/HDD, on board M.2 slot, support dual width FHFL GPU. ●Support Intel 8/9th Core i7/i5/i3 (Coffee Lake-S) series processor, LGA 1151 ●2x channel DDR4 2400MHz un-buffered / Non-ECC SODIMM, max support 32GB ●1x dual width PCIe Gen 3 x16 slot, supporting RTX-3070

這篇文章 AEWIN的D8AI NLP Document Recognizer解決方案 最早出現於 AEWIN。

AEWIN partners with D8AI to bring you AI powered Chatbot solution. D8AI was founded by industry veterans aiming to bring AI into everyday applications through innovations with in-house AI engines that empower solutions across their portfolio. The Chatbot solution is designed as enterprise-grade customer care AI that can provide relevant and timely assistance or response. It can integrate into many existing messaging and social media platforms, with a single unified management backend. This allows easy management and monitoring of user interactions.

The AI model can be easily trained, modified, and retrained through a simple and intuitive webGUI. The data import for training is done through an easy interface, and Q&A specific for the application can be loaded into the system with simple Excel file. Chatbot backend interface will provide checks of responses for validity, with additional ranked responses that the manager can choose for guided training when chatbot responses can be improved. Customer care personnel can monitor and intervene at any moment transparently to the end user in cases that could not be handled by the AI. Data from these interventions can load back into the AI training regimen and update the AI model at any time, as well as performing big-data analytics to track trending topics and more.

Additional services and modules can be added to tailor Chatbot system for your specific use cases. The AI is date aware, and customized greetings and responses can be taught to the AI for seasonally relevant responses. There are options to add scripts for general chat interactions for more natural conversation flow. The AI is aware of tone and emotional responses which then triggers alert to the manager for intervention to de-escalate the situation or provide more appropriate replies.

AEWIN computing platforms has been validated by D8AI and are the ideal hosts for D8AI’s suite of innovative software solutions. AEWIN offers system equipped with the latest 3rd Generation Xeon-SP or Zen3 powered EPYC 7003 series processors. SCB-1932 and SCB-1937 are sister platforms, powered by Xeon-SP and EPYC processors respectively with acceleration provided by up to 2x dual-width NVIDIA Datacenter accelerators. These are powerful system capable of handling thousands of concurrent users, enabling the chatbot to service the needs of the largest organizations. For more modest user count, the SCB-1826 and SCB-1833, powered by workstation class CPU by Intel and AMD, are cost effective alternatives.

AEWIN has been steadfast in our pursue in bringing highly configurable platforms for the emerging edge AI market. Our systems have been chosen by many ISV as the foundation to host their latest software innovations. Our deep partnership with D8AI allowed us profound insight on the specific workload, which enable us to tailor BIOS and firmware settings for optimized results, as well as carefully chosen hardware specifications tuned for exacting needs.

SCB-1932 Dual 3rd Gen Intel Xeon SP High Performance Edge Computing system 8x AEWIN Expansion Module bays, supporting Network Expansion Modules, GPU, U.2 Convertors ● Support 3rd Gen Intel Xeon Scalable Processor(Ice Lake-SP) ● Support Max 2TB DDR4 3200MHz & Optane DC Persistent Memory ● Max support 8x PCIe Gen4 x8 slots for Network Expansion Module ● Support BMC/IPMI& hardware One-Key Bypass function

這篇文章 AEWIN的D8AI Chatbot解決方案 最早出現於 AEWIN。