Bringing AI to the Edge- Application Cases

MIC-720AI leverages AI inference technology to perform traffic monitoring on massive amounts of collected data; surpassing traditional vehicle recognition methods used for object tracking.

MIC-720AI fulfills deep learning computing requirements at the roadside, where metadata is packaged and transmitted to the central control room. It also provides multiple interfaces to integrate with other traffic equipment.

Taipei’s traffic lights’ lengthy countdowns exist to provide fixed and enough time for pedestrians to safely cross wide roads. Such arrangements burdensome around midnight when fewer pedestrians and vehicles traverse Taipei’s streets.

After deploying this AI system, night wait times at red lights on arterial roads decreased by 35% and arterial roads green light periods increased by 7 ~ 79%. Also, it reduces carbon dioxide emissions by around 23 tons every year and results in an annual economic benefits of around TWD 1.83 million for each intersection.

In recent years, beekeepers have been reporting they are losing, on average, 30 percent of all honeybee colonies each winter. This is twice the loss considered economically tolerable. Similarly, wild bee populations are also in decline.

So, we must learn more about bee hive health, bee behavior, and queen failure in an effort to limit future bee hive loss. Now, to get a complete picture of the health of the bee hive, SAS had to be able to collect, visualize a variety of IoT data and use AI analyze from video and audio data.

To maintain consistent quality for bakeries, MIC-720AI deploys various AI models to ensure that all cookies are well baked in the production line. Compared to traditional quality control procedures performed through visual inspection, AI inference detects subtle levels of difference in how individual cookies are baked, adjusts the oven to meet quality standards, and avoids under-baked or overbaked products.

AI inference requires high computation and needs a GPU-based solution to accelerate the computation. MIC-730AI is powered by an NVIDIA® Jetson AGX Xavier™ GPU and is employed as the edge AI system. MIC-730AI's great processing power makes it possible to automatically inspect high-precision textile goods with better speed and accuracy.

Textile defect inspections using AI inference technology can efficiently identify the most subtle defects and ensure high quality products.

MIC-730AI conducts AI image analysis to compare and identify heat sinks with appearance defects.This is made possible by feeding AI model training system with images of defective products that collected from clients, prior to the import of software and hardware. After AI model completes relevant training, the trained models are placed into the MIC-730AI.

AI visual inspection can then be conducted to evaluate the flatness and identify crushes, stains, scratches and other defects that are difficult to categorize through general physical rules.

The Human Factor Information Datamation AI System uses cameras to record worker actions and then analyze images and video through AI inference. This system assists foundries in identifying bottlenecks and improving production efficiency.

Five major electronic manufacturing service foundries adopt this systems. With the aid of AI deep-learning software, factories increased their Unit Per Hour (UPH) by 5% within only two months.

By using Edge AI, video streams from regular cameras can be analyzed in real time at the network edge and used to make critical security decisions. They can detect if triggered events are true intrusions or just false alarms, and they can instantly send notifications directly to your mobile phone with recorded video and pictures. Best of all, there is no additional cost on a security system, no expensive installation of sensors, and no high maintenance costs, this is the perfect solution for using your pre-installed security cameras to keep your family and assets safe.

Modern construction sites utilize real-time visual detection systems that analyze 20 to 30 live streams simultaneously. In these systems, visual AI increases the visibility of on-site workers and equipment to improve responsiveness to potentially dangerous situations — such as an employee carelessly approaching moving machinery. Real-time video feeds are available via the cloud to any device with a screen. To avoid tragedies or accidents, real-time alerts are sent to screens; and via methods including SMS and email.