Elevator door state detection method and system based on image histogram

Abstract

The invention discloses an elevator door state detection method based on an image histogram. The method comprises the following steps of S1, collecting video image information in a lift car; S2, according to the collected video image information, performing analysis, judging whether a currently-monitored elevator door of an elevator car is in the opening state or not, and transmitting the data toa data analysis unit; S3, enabling a door state detection unit to register a background template in an H channel of an HSV space under the state that the door is completely closed according to the detection area transmitted by a video collection module, continuously calculating the difference between a current frame histogram and a background histogram subsequently, and judging the opening and closing state of the door.

Description

technical field [0001] The invention belongs to the technical field of elevator safety, and in particular relates to an elevator door state detection method and system based on an image histogram. Background technique [0002] At present, the elevator has become the standard equipment in many buildings in the city, and it is an important means of transportation for people in modern life. Due to its frequent use and its own airtight characteristics, the elevator environment has a strong demand for image monitoring. The traditional elevator monitoring only has the basic video recording function, and cannot intelligently judge the current state of the elevator door. When the elevator fails suddenly, the community management personnel can only query the monitoring images laggingly, and the crisis handling mechanism responds slowly and has certain hysteresis. [0003] Chinese invention patent application CN106204659A discloses an elevator door opening and closing detection meth...

AI Technical Summary

Problems solved by technology

There are following deficiencies in this technical solution: (1) It is necessary to extract corner points from the four sub-regions of the region of interest at the same time, and perform optical flow tracking respectively, and count the direction of motion

The whole process is slightly cumbersome and the amount of calculation is large

(2) The local image feature detection method using optical flow tracking corner points is greatly affected by changes in external lighting. Once the lighting of the elevator car itself changes greatly, or the external lighting interference caused by opening and closing doors, the tracking algorithm will fail.

There are the following deficiencies in the technology of this solution: (1) This solution can only use depth cameras as monitoring equipment, and the cost of depth cameras is relatively high

Images