When it comes to supporting users in building feature-rich custom enterprise applications, Microsoft has always been a step ahead. Recently, the tech giant added two new tools to its Azure AI cloud suite of products, which will enable developers to add richer artificial intelligence capabilities into the apps they develop. According to Microsoft, the new additions to its Azure Cognitive Services will make it easy for developers and data scientists to deploy, manage, and secure AI functions directly into their applications.
In his latest Azure Blog, Anand Raman (Azure AI Platform Product Manager) shares details about both the updates:
Azure Custom Vision – general availability of Custom Vision to more accurately identify objects in images
Anomaly Detector Service – a preview of the new Anomaly Detector Service which uses AI to identify problems so companies can minimize loss and customer impact
Azure Custom Vision
Custom Vision is powered by automated machine learning, which makes it easy and fast for developers to build, deploy, and improve custom image classifiers to quickly recognize content in imagery. Developers can either train their own classifier to identify what’s important in a given scenario, or they can export custom classifiers to run them offline and in real-time. In real-time, custom classifiers can be run on iOS (supported by CoreML) and Android (using TensorFlow) and many other devices on the edge.
Custom Vision offers the following advantages:
Optimize exported models to meet the constraints of mobile devices so that the models provide unbelievable and highly accurate outputs. That means, you can train the model in Cloud to detect
unique objects from images and run the model anywhere.
Easy for developers to integrate computer vision capabilities into applications with 3.0 REST APIs and SDKs. Custom Vision supports iterative improvement of models, so you can quickly train a
model, prototype in real world scenario, and use the resulting data to enhance the model.
Advanced training with a new machine learning backend for improved performance. This works well on challenging datasets and fine-grained classification so that you can develop high quality
Anomaly Detection Service
Anomaly Detector arrives as a new Cognitive Service that allows you to detect unusual patterns or rare events in your data that could translate to identifying business problems. This new AI service can automatically surface incidents as soon as they happen.
Currently, many teams across Azure and other core Microsoft products are depending on Anomaly Detector to increase the reliability of their systems. It lets you detect irregularities in real-time and accelerate troubleshooting. Anomaly Detector helps in ensuring high data accuracy and developers can easily embed anomaly detection capabilities into their applications using a single API.
Some of the common use cases include identifying business incidents and text errors, monitoring IoT device traffic, detecting fraud, and responding to changing markets. It’s highly useful for content providers as they can use Anomaly Detector to automatically scan video performance data specific to a customer’s KPIs. Similarly, video streaming platforms can use Anomaly Detector to scan millions of video data sets. For content providers, even a missed second in a video can lead to significant revenue loss. With the new AI service, tracking metrics across millions of data sets is quite easy.
Microsoft continues its efforts to make Azure the best place to build AI. Undoubtedly, both the updates are crucial milestones in their journey towards this goal.