A work analysis and improvement tool that combines AI (machine learning) and skeletal detection technology for visualizing and monitoring human tasks such as assembly work.
The Impulse work analysis and improvement tool captures human work using a camera and automatically analyzes the quality, efficiency, and load of the work by recognizing information such as the movements of objects and human skeletal structures with AI.
Impulse was released in 2014 ahead of the market with the aim of practical application of machine learning technology, and it is an anomaly detection solution that has been "refined in the field" by numerous customers.
In addition to having data scientists with advanced analytical skills, it aims to be a platform that can be used by a wider range of users, equipped with architecture and features that facilitate the easy introduction and operation of machine learning technology. Its most notable feature, auto-modeling (Patent No. 6315528), automatically classifies the characteristics of complex and vast data and provides functionality to derive the optimal analysis model through simulations using standard algorithms.
<Data Analysis Process>
- Data organization
- Policy determination
- Feature selection
- Tuning
- Model operation
