Method and system for identifying reading of pointer instrument in inspection scene

Abstract

The invention discloses a pointer instrument reading identification method and system in an inspection scene. The method comprises the following steps: acquiring a pointer instrument image; positioning a pointer and an instrument in the pointer instrument image based on the positioning detection model, and positioning scale readings of the instrument in the image; based on the positioned data, fitting a dial plate image in the pointer instrument image by adopting an OCR direction detection model; performing correction operation on the dial plate image based on plane rotation and space overturning to obtain a corrected dial plate image; carrying out pointer straight line fitting and scale reading number identification based on the corrected dial plate image, wherein an OCR identification model is adopted in the number identification; on the basis of instrument prior information, a digital recognition result is calibrated; and obtaining the instrument reading based on the calibration result and the pointer straight line fitting result. The instrument can be accurately positioned and corrected, and the success rate and accuracy of final reading of the pointer instrument are guaranteed.

Description

technical field [0001] The invention belongs to the field of intelligent robots, and in particular relates to a method and system for recognizing pointer instrument readings in a patrol inspection scene. Background technique [0002] As important indicating instruments and instruments, pointer instruments are widely used in industrial fields, such as pressure gauges, voltmeters, ammeters, thermometers, etc. These instruments are used in industrial sites to measure and indicate on-site environmental parameters, production index parameters, and equipment operating parameters. The reading and recording of the values ​​of these meters, ie meter reading, is an essential part of the production process. In the traditional industrial production site, the value reading of the instrument is transcribed by workers going to the site, which requires a lot of manpower to complete this task. Moreover, in some complex environments and dangerous production situations, it is more difficult ...

AI Technical Summary

Problems solved by technology

The patent text (CN111598109A) provides an intelligent recognition method for substation pointer instrument readings, which uses traditional image processing techniques (including image correction, grayscale, edge detection, circle transformation detection dial, Hough line detection and detection pointer) to obtain instrument representation number; the patent text (CN111931776 A) provides a two-pointer meter reading method based on deep learning, uses the deep learning method to locate the meter, then performs Hough circle detection on the meter image, and simultaneously performs ellipse connected domain screening, and combines the two methods to obtain The circular dial of the instrument, the method is limited by the Hough circle detection method is not suitable for the target imaging rotation and flip scene, the actual application is not good

[0004] To sum up, the existing methods for automatic recognition of pointer instrument indications have certain defects. In the actual machine room inspection application, it is easy to recognize the target image due to different lighting, angles, and postures, and it is difficult to recognize when the target image is partially occluded or the image is blurred. It is almost difficult to effectively identify under conditions such as image tilt and image tilt, so it is difficult to meet the requirements of practical applications

Images