Visualization of Time Domain Signals and Signal Cognition
Yuji Kimura, Seiji Hayano, Yoshifuru SAITO, Kiyoshi HORII · Journal of the Visualization Society of Japan · 2004
Previously, we have proposed a full automatic metallic cans classification device along with the magnetic sensor signal visualization.In this paper, we have carried out more practical experiments along with this device. In order to work out a practically usable full automatic metallic cans classification device, it should be considered that the target cans should be processed with more degree of input freedom, i. e., free falling cans should be cognized and classified in the device. To overcome this difficulty, we have exploited a multi-stage magnetic sensor system and carried out a sequential systematic experimental works.As a result, it is revealed that simple extension from the previous system to the multi-sensor system could be cognized and classified the tested target cans in a most high accuracy. Thus, we have succeeded in developing the second stage of the full automatic metallic cans classification device.