Multi-pointer instrument alarm detection method and device, computer equipment and storage medium

Abstract

The invention relates to a multi-pointer instrument alarm detection method and device, computer equipment and a storage medium. The method comprises steps of obtaining an image of an instrument so asto obtain an initial image; inputting the information into a target detection model for identification to obtain a target detection result; verifying the target detection result, and cutting the initial image to obtain a high-definition pointer instrument picture; inputting the high-definition pointer instrument picture into an instance segmentation model for instance segmentation to obtain different state region segmentation results; performing main color identification on different state area segmentation results, and dividing a safety area and an alarm area in the state area according to the configuration information; inputting the high-definition pointer instrument picture into a pointer identification model for pointer position identification to obtain an identification result; acquiring the pointer information according to the identification result; detecting a state of the pointer, and generating the notification information according to the state of the pointer. According to the method, safety of the area where the pointer is located is accurately predicted, accuracy is high, and multi-pointer recognition is supported.

Description

technical field [0001] The invention relates to a detection method of a pointer instrument, more specifically to a multi-pointer instrument alarm detection method, device, computer equipment and storage medium. Background technique [0002] At present, most of the safety monitoring instruments installed by production enterprises are mechanical pointer instruments. It is difficult to directly connect to the early warning system of the Internet of Things during the transformation of the Internet of Things for safety production. The more mainstream processing method in the market now is the way of direct meter reading by camera, that is, taking pictures of the instrument panel through the camera, and then reading the value displayed on the pointer of the meter through image recognition technology, and judging whether to trigger an early warning event based on the value. But the recognition rate of this method is relatively low, and the versatility is also poor. [0003] Chines...

AI Technical Summary

Problems solved by technology

However, the recognition rate of this method is relatively low, and the versatility is also poor.

[0003] Chinese patent CN201910294823.2 provides an image recognition-based pointer instrument early warning method, which is to detect the edge contours of all objects by converting the image into a binary image to obtain an edge contour map; find the contour through a straight line detection algorithm Figure all the straight lines; use the constraints to get the pointer of the instrument and judge whether the straight line is within the range of the warning area, but this method has a poor detection effect on thin pointers, and can only support single pointer recognition, and there are reflections and shadows on the dial. There is a big discount, the accuracy rate is not high

Images