• Journal of the Korea Society of Computer and Information
• /
• v.23 no.2
• /
• pp.27-34
• /
• 2018

In this paper, we propose a new method to read the hand of analog gauges to prepare for the smart factory. In addition, we suggest a new and improved method that can apply, in general, diverse analog gauges even if their scale types and ranges are various. Many companies are making great efforts to build smart factories that increase energy efficiency and automation. Managers use a variety of equipment and tools to manage the production process at the factory. In this kind of factory, analog gauges have been often used with many equipment and tools. Analog gauges are mostly circular in shape, and most papers use circular hough transform to find the center and radius of a circle. However, when the object to be found is not of the correct circle type, it takes a long time to recognize the circle using the circular hough transform, and the center and radius of the circle can not be calculated accurately. The proposed method was tested on various circular analog gauges. As a result, we confirmed that our method is outstanding.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.29 no.4
• /
• pp.389-394
• /
• 2023

In recent years, use of machine automation is rising in the industry. Ships also obtain machine condition information from sensor as digital information. However, on ships, crew members regularly surveil the engine room to check the condition of equipment and their information through analog gauges. This is a time-consuming and tedious process and poses safety risks to the crew while on surveillance. To address this, engine room surveillance using an autonomous mobile robot is being actively explored as a solution because it can reduce time, costs, and the safety risks for crew. Analog gauge reading using an autonomous mobile robot requires digitization for the robot to recognize the gauge value. In this study, image processing techniques were applied to achieve this. Analog gauge images were subjected to image preprocessing to remove noise and highlight their features. The center point, indicator point, minimum value and maximum value of the analog gauge were detected through image processing. Through the straight line connecting these points, the angle from the minimum value to the maximum value and the angle from the minimum value to indicator point were obtained. The obtained angle is digitized as the value currently indicated by the analog gauge through a formula. It was confirmed from the experiments that the digitization of the analog gauge using image processing was successful, indicating the equivalent current value shown by the gauge. When applied to surveillance robots, this algorithm can minimize safety risks and time and opportunity costs of crew members for engine room surveillance.