Archive dense cabinet self-certification method based on artificial intelligence

Abstract

The invention provides an archive dense cabinet self-certification method based on artificial intelligence. The method comprises steps that a key point detection network processes a dense cabinet image to obtain a crank handle key point heat map, the dense cabinet image of the crank-handle-free dense cabinet is screened according to the crank handle key point heat map, and a cabinet body detectionnetwork processes the dense cabinet image of the crank-handle-free dense cabinet to obtain a cabinet body surrounding frame; a crank handle key point or the central point of the cabinet body surrounding frame is projected into the BIM to obtain a three-dimensional projection point, whether the dense cabinet is opened or not is judged according to coordinates of the three-dimensional projection point, and after the dense cabinet is opened, whether a worker has an unauthorized opening behavior or not for the dense cabinet with the crank handle is judged; for a serried cabinet without a rockinghandle, whether the worker has an unauthorized opening behavior or not is judged, and whether the serried cabinet is damaged or not further needs to be judged. According to the method, when the densecabinet is opened, whether related personnel have permission to open some dense cabinets or not can be judged, and safety of archive information in the dense cabinet is ensured.

Description

Technical field [0001] The present invention relates to the field of artificial intelligence, in particular, it is based on artificial intelligence intensive archives cabinet witness method. Background technique [0002] For some open or semi-open Shelves, relevant personnel can not know in real time whether someone opened not within its purview Shelves, archives will bring great losses, if to replace all the intelligent Shelves, We need to spend a lot of time and capital. [0003] Fault detection conventional electric filing cabinet is often found and reported by the staff to repair, and moves the error caused by the mechanical structure often requires re-calibration, but the body is difficult to recognize the error in the distance, i.e. partial failure filing cabinet body difficult to accurately identify and discovery. Inventive content [0004] To solve the above problems, the present invention provides a counter-based artificial intelligence intensive archives witness, the m...

AI Technical Summary

Problems solved by technology

[0002] For some open or semi-open shelves, the relevant personnel cannot know in real time whether someone has opened the shelves that are not within their authority, which will bring huge losses to the archives. If all the shelves are replaced with smart shelves, cost a lot of capital and time

[0003] The fault detection of existing electric dense cabinets is often discovered by the staff and reported for repair, and the error caused by the mechanical structure often needs to be recalibrated when moving, but it is difficult for the human body to recognize the error in the distance, that is, the partial failure of the dense cabinet. Difficult to accurately identify and find

Images