A Novel Gas Meter Verification Method via Adaptive Template Matching and Pulse Activation Decision

In the industrial field, many researchers utilize the machine vision algorithm to verify the gas meter. This paper put forwards an efficient gas meter verification approach, which is consists of the core gear location and gear overlap judgment. First, due to the uncertainty of size, position, and shape of a gear, an adaptive template matching method is proposed to solve the problems of the size variation and location uncertainty. Meantime, because the gear owns the centrosymmetric characteristic, a maximum connected domain strategy is utilized to confirm the initial rotation location of the gear. And a pulse activation decision strategy to measure the gear overlap. Then, we employ the error analyzer of the gas meter to assess the performance of our algorithm. For verification purposes, the proposed algorithm has conducted a series of experiments in the cooperating factories to demonstrate the effectiveness and accuracy of the entire scheme.