IBM Maximo Visual Inspection overview

IBM® Maximo® Visual Inspection platform, built on cognitive infrastructure, is a new generation of video/image analysis platforms. The platform offers built-in deep learning models that learn to analyze images and video streams for classification and object detection.

IBM Maximo Visual Inspection includes tools and interfaces for anyone with limited skills in deep learning technologies. You can use IBM Maximo Visual Inspection to easily label images and videos that can be used to train and validate a model. The model can then be validated and deployed in customized solutions that demand image classification and object detection.

The following are the main features of IBM Maximo Visual Inspection:
Streamlined model training
You can use existing models that are already trained as starting point to reduce the time required to train models and improve trained results.
Single-click model deployment
After you create a training model, you can deploy an API with one click. You can then develop applications based on the model that you deployed.
Data set management and labeling
You can manage both raw and labeled data.
Video object detection and labeling assistance
Videos that you import can be scanned for objects and the objects can be automatically labeled.

Architecture overview

The architecture of IBM Maximo Visual Inspection consists of hardware, resource management, deep learning computation, service management, and application service layers. Each layer is built around industry-standard technologies.

This image illustrates the architecture layers with Vision recognition service on the top, the service management layer next, the computation layer below that, followed by the resource management layer, then the components: POWER, Accelerator, Data store, and Network on the bottom.

Table 1. Overview of the architecture layers
Architectural Layer Description
Infrastructure Layer Consists of hardware systems that support IBM Maximo Visual Inspection, including virtual machines (containers), accelerators (GPUs), storage systems, networks, and so on.
Resource Management Layer Coordinates and schedules all computing resources.
Deep Learning Calculation Layer Consists of deep learning algorithms, including data processing modules, model training modules, and prediction modules.
Service Management Layer Manages user projects in a graphical interface, including image preprocessing, data annotation management, data set management, training task management, model management, and API management.
Application Service Layer Located on the top of the IBM Maximo Visual Inspection platform, it is responsible for managing all application-related services, including image labeling and preprocessing services, video annotation services, customized image classification services, and customized object detection services.
Note: IBM Maximo Visual Inspection was formerly named IBM Visual Insights (1.2.0) and, prior to that, IBM PowerAI Vision (1.1.5 and earlier).