Fujitsu One Shot Machine Learning Tech Delivers Results with Limited Training Data
Fujitsu today announced the development of a technology that automatically detects different work elements—like, removing a part, tightening a screw, or installing a cover—from video data consisting of a series of tasks from manufacturing lines and other places of manual work. Fujitsu has successfully developed a technology to automatically detect work elements even for different video data of the same work, taking into account variations in motion and other similar movements, by learning each work element using data obtained by dividing a single video for each element as training data.
Fujitsu has expanded upon its Actlyzer technology for detecting human movement and behaviour from video data, and developed an AI model that takes into account the variation of each movement and the difference in individual workers’ movements, using work data from one person and the data divided by each work element as training data.
When this technology was applied at the Yamanashi Plant of Fujitsu I-Network Systems, which manufactures network equipment products, Fujitsu applied the developed technology to analyze the following three work processes, parts setting, assembly, and visual inspection, and conducted an evaluation. In each process, by only training with the division data of each work element of one person’s video, the work element is automatically detected even in another video of the same work with 90% or more accuracy. As a result, the cycle of kaizen activities can be repeated more frequently, helping to improve work efficiency and accelerate the transfer of skills.
By leveraging this technology, Fujitsu will continue to contribute to the promotion of work process improvement activities and the passing on of specific skills for more efficient work at various sites. In addition to the manufacturing industry, Fujitsu will continue to verify this technology in various use case scenarios, including logistics, agriculture, and medical care, with the aim of putting it into practical use within fiscal 2021
