Accurate counting of the number of images with AI image processing - determining the excess or shortage of sheet-like materials and 0.05mm thin products in as little as 0.3 seconds using deep learning.
ai-Numbers analyzes sheet products with complex edge surfaces and significant cross-sectional variations from the side at high resolution using AI, automatically determining excess and deficiency in as little as 0.3 seconds (※). Furthermore, through a unique learning method that eliminates the need for threshold settings, it reduces setup time while fundamentally eliminating counting errors, thereby alleviating the workload in manufacturing environments.
(※) Depends on the field of view and counting conditions.
Conventional image counting machines extracted one-dimensional brightness from images and counted changes in brightness. As a result, they could not accurately count due to changes in the state of edge surfaces or because items in transparent bags were partially visible, leading to inaccuracies.
ai-Numbers, through deep learning, detects two-dimensional features of edge surfaces (color, patterns, texture), allowing for accurate counting of sheet products with complex edge surfaces and significant cross-sectional variations.
It supports the automation of quality control and the reduction of personnel on production lines, contributing to manufacturing DX solutions, smart factory initiatives, and cost optimization.