THE DEVELOPMENT OF TIDE GATE MONITORING SYSTEM USING IMAGE RECOGNITION AI

Abstract

 Due to giant typhoons in recent years, the tide gate open-close status confirming system is urgently required to alleviate storm surge damage, and to take immediate and wide-ranging evacuation activities and emergency measures. This paper describes the prototype of the system and the demonstration test at a fishing port. This system detects of gate status by image recognition AI, and transmit the status and information of peripheral state.

 As a result of demonstration test, the quality of the model was confirmed with 92.2% correct status designations. And the communication situation of LPWA which used for power saving was good. This system was received a good evaluation from administrator of the tide gate in the hearing after the experiment.