A vision-based measurement model for instruments without internet connectivity. (October 2018)

• Record Type: Journal Article
• Authors: Bojaca, Edwin; Anzola, John; Jiménez, Andrés; García-Díaz, Vicente
• Is Part Of: Computers & electrical engineering. Volume 71(2018)
• Page Start: 533
• Page End: 545
• Publication Date: 2018-10
• Subjects: Artificial-vision -- Industrial-instrumentation -- Image-processing -- Tesseract-architecture -- Vision-based measurements -- Internet of things -- Measurement model -- Raspberry Pi -- Industrial indicators; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2018.08.009
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: The camera's angle of inclination in the proposed model for the analog and digital instruments, affects the measure extracted from the image. The number of digits classified positively in digital instruments is highly dependent on the position of the image capture, mainly for perspective and distortion errors. This model obtains values with an error of 0.00013 for analogous indicators, taking an image at 90 degrees with a deviation of 10 degrees. This model obtains values with zero error for digital instruments at 90 degrees with a deviation of 10 degrees, based on the Tesseract architecture. Mechanical operating instruments have reading errors due to the slow change in digits, while electronic operating instruments have reading errors due to the high speed of change between the digits. Abstract: The application of artificial vision to the field of instrumentation and measurement, known as vision-based measurement, is rapidly expanding due to recent developments in the Internet of Things (IoT). In this article, a model for obtaining instrument measurements from analog or digital instruments with no Internet connectivity using digital processing of images and the Tesseract architecture is demonstrated and implemented in a Raspberry Pi. The model's implementation is used to create an interoperability instrument based on artificial vision that transforms the indicator's visual information into digital data. The model is evaluated by capturing images of the indicator or instrument at 30, 60, 90, 120, and 150 degrees, with a 30 cm distance from a camera connected to a Raspberry Pi. The influence of the image's capture angle is analyzed in the proposed model, extending the model's evaluation into other industrial indicators of measurement.