Digitalization of 32-Channel Manometer Data with Image Processing
This project introduces an innovative image processing system that enables the digital transformation of traditional analog multi-channel manometers. In particular, the manual reading of 32-channel manometers is a time-consuming process, prone to human error, and unsuitable for digital integration. The developed system simultaneously captures all channels with a single camera and uses computer vision algorithms to detect fluid levels, converting them into real-time digital values. Thus, measurement data can be automatically recorded, analyzed, and accessed remotely. The project provides a low-cost and scalable innovation that can be used both in academic laboratories and in industrial testing systems.
The main objective of the project is to facilitate the integration of analog measuring devices into digital systems and to eliminate errors caused by manual processes. Today, multi-channel pressure measurements are generally performed either by using a separate sensor for each channel or by manual observation. Sensor-based solutions require high costs and complex infrastructure, while manual reading methods are slow and error-prone. This invention fills an important gap in both literature and industry by enabling the simultaneous and reliable digitization of 32 channels through a camera- and software-based approach. Its innovative aspect lies in minimizing the need for expensive hardware and adapting image processing and artificial intelligence algorithms to multi-channel manometers, thereby integrating traditional devices into Industry 4.0 infrastructures.
There are various studies in the literature on digitizing analog gauges. Since the early 2010s, basic image processing algorithms have been applied to fluid level measurement, and after 2015, AI- and CNN-based solutions have been developed. From 2020 onwards, IoT-based pressure sensors came to the forefront, but these solutions are expensive and generally suitable for a limited number of channels. After 2022, AI-supported smart measurement systems have been developed, but they mostly focused on 1–16 channels. Academic research has particularly concentrated on the automatic reading of single analog gauges (dial-type manometers) and the use of synthetic data-based methods. However, there is no comprehensive study in the literature addressing the digitization of large-scale, multi-channel manometers—such as 32 channels—using image processing. This highlights the originality and research contribution of the project.
This project fills the gap in the literature regarding the digitization of multi-channel manometers and offers a practical solution for industry. Its most important contribution is the ability to read 32 channels simultaneously with a single camera. This significantly reduces measurement time, eliminates human error, and increases the reliability of the data. Moreover, since there is no need to use separate sensors for each channel, the system is cost-effective, easy to install, and scalable. The immediate recording of digital data, along with remote access and automatic reporting, meets modern industrial requirements. Therefore, the project stands out as a digital innovation product compatible with Industry 4.0 that can be used both in academic research and in industrial applications.
