Convolutional Neural Networks for Automatic Meter Reading
| Autor: | Rayson Laroca, Matheus Alves Diniz, William Robson Schwartz, Victor Barroso, Gabriel Resende Gonçalves, David Menotti |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
FOS: Computer and information sciences
Artificial neural network Computer science Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition Image processing 02 engineering and technology Image segmentation Optical character recognition computer.software_genre Convolutional neural network Atomic and Molecular Physics and Optics Computer Science Applications Data modeling 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Data mining Electrical and Electronic Engineering computer Protocol (object-oriented programming) Automatic meter reading |
| Popis: | In this paper, 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 new public dataset, called UFPR-AMR dataset, with 2,000 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 less than 200 images for training. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|