PP:Classification of Time-Series Images Using Deep Convolutional Neural Networks

The 10th international conference on machine vision; C类

Methodology: 非主流方法

2 stages:

1. convert time series data to recurrence plot. 数值*时间长度----------> 时间长度*时间长度.

2. fed into CNN model.

潜在问题:

1. 由time series data 转化成为 recurrence plot是否丢失了信息,丢失了哪些信息------未知

2. cnn分类效果是否比别的好. 文章在在20个数据库上进行了测试,试验结果并没有很明显的提高.

PP:Classification of Time-Series Images Using Deep Convolutional Neural Networks

Supplementary knowledge:

1. recurrence plot