Intelligent reading recognition method of a pointer meter based on deep learning in a real environment. (1st May 2022)

• Record Type: Journal Article
• Authors: Zhou, Dengke; Yang, Ying; Zhu, Jie; Wang, Ku
• Is Part Of: Measurement science & technology. Volume 33:Number 5(2022)
• Publication Date: 2022-05-01
• Subjects: pointer meter -- intelligent reading -- deep learning -- local angle method; Physical measurements -- Periodicals; Scientific apparatus and instruments -- Periodicals; Equipment and Supplies -- Periodicals; Science -- instrumentation -- Periodicals; Technology -- instrumentation -- Periodicals; Mesures physiques -- Périodiques; Physical measurements; Scientific apparatus and instruments; Periodicals; 502.87
• DOI: 10.1088/1361-6501/ac4079
• Languages: English
• ISSNs: 0957-0233

Abstract

Abstract: Accurate reading of a pointer meter is a crucial task in complex environments, such as substations, the military and aerospace. Current recognition algorithms are mainly used to identify the same type and a non-tilt meter, which has limited application in a real environment. In this paper, we propose a novel end-to-end intelligent reading method for a pointer meter based on deep learning, which locates the meter and extracts the pointer simultaneously without any prior information. In particular, the pointer is directly and precisely extracted using the designed semi-pointer detection method without any handcrafted features designed in advance, which avoids accumulated error caused by preprocessing. Based on the extracted panel object, including the semi-pointer, panel center and scale characters, the indicated value of the pointer is obtained by a local angle method, which achieves better performance than the traditional angle method by referring to the neighboring scale lines of the pointer. Experimental results demonstrate that the method is faster and more effective than some common methods. It is worth noting that this study has the advantage of being able to recognize pointer meters under complex conditions such as tilt, rotation, blur and illumination. This is acceptable for the actual application requirements in a real environment with a recognition accuracy of 99.20% and an average reference error of 0.34%.