Convolutional neural networks for automatic meter reading
Convolutional neural networks for automatic meter reading
复制标题
DOI：
10.1117/1.jei.28.1.013023
复制
发表时间：
2019-01-01
影响因子：
1.1
通讯作者：
Menotti, David
中科院分区：
文献类型：
--
作者：
Laroca, Rayson;Barroso, Victor;Menotti, David
我们通过利用卷积神经网络（CNN）的高性能来解决自动抄表阅读（AMR）问题。我们设计了一个两阶段的方法，采用Fast-YOLO对象检测器进行计数器检测，并评估了三种不同的基于CNN的计数器识别方法。在AMR文献中，由于图像属于服务公司，因此研究界无法获得大多数数据集。在这个意义上，我们引入了一个公共数据集，称为联邦巴拉那大学AMR数据集，有2000个完全和手动注释的图像。据我们所知，该数据集比文献中发现的最大公共数据集大三倍，并且包含一个定义明确的评估方案，以协助AMR方法的开发和评估。此外，我们建议使用数据增强技术来生成具有更多示例的平衡训练集，以训练CNN模型进行计数器识别。在所提出的数据集中，获得了令人印象深刻的结果，并对每个模型进行了详细的速度/精度权衡评估。在一个公共数据集中，最先进的结果是使用
We tackle automatic meter reading (AMR) by leveraging the high capability of convolutional neural networks (CNNs). We design a two-stage approach that employs the Fast-YOLO object detector for counter detection and evaluates three different CNN-based approaches for counter recognition. In the AMR literature, most datasets are not available to the research community since the images belong to a service company. In this sense, we introduce a public dataset, called Federal University of Parana-AMR dataset, with 2000 fully and manually annotated images. This dataset is, to the best of our knowledge, three times larger than the largest public dataset found in the literature and contains a well-defined evaluation protocol to assist the development and evaluation of AMR methods. Furthermore, we propose the use of a data augmentation technique to generate a balanced training set with many more examples to train the CNN models for counter recognition. In the proposed dataset, impressive results were obtained and a detailed speed/accuracy trade-off evaluation of each model was performed. In a public dataset, state-of-the-art results were achieved using