Abstract: Screw hoist is widely used in water conservancy projects，however，the unique structure of screw hoist is prone to anti-top gate accidents in case of gate falling blocked and not timely shutdown，bringing serious security risks.In this regard，an intelligent anti-top gate control scheme for screw hoist based on machine vision was proposed.Firstly，the operation characteristics of screw under different working conditions were analyzed.Secondly，based on the high-definition camera next to the hoist，the machine vision algorithm was used to detect the position and speed of the screw，analyzed the motion state of the gate in real time，respond to the abnormal situation in time and stop the machine.This scheme improves the utilization rate of the equipment.Compared with the traditional shaft pin load protection scheme，it has the characteristics of flexible deployment of non-contact detection and no additional installation of weighing sensors.The research results can provide a new idea for the control scheme of traditional electromechanical equipment such as screw hoist under the background of informatization.