Abstract
A small target detection method based on deep learning is proposed. First, random background parts are sampled from some cloud-sky images. Then, random generated target spots are added to the backgrounds with controlled signal to background noise ratio (SNR) to generate target samples. Then training and testing results show that the performance of deep nets is superior to tradition small target detection techniques and the selection of sampling SNRhas an important effect on nets training performances. SNR = 1 is a good selection for deep nets training, not onlyfor small target detection,but also for other applications.
| Original language | English |
|---|---|
| Title of host publication | Current Trends in Computer Science and Mechanical Automation Vol.1 |
| Subtitle of host publication | Selected Papers from CSMA2016 |
| Publisher | de Gruyter |
| Pages | 211-220 |
| Number of pages | 10 |
| ISBN (Electronic) | 9783110584974 |
| ISBN (Print) | 9783110584967 |
| Publication status | Published - 9 Jan 2018 |
Keywords
- Deep learning
- Nerual Network
- SNR control
- Small target detection