Multi-Attention Object Detection Model in Remote Sensing Images Based on Multi-Scale

[업데이트 2019.08.23 18:45]

 

1. 논문

Multi-Attention Object Detection Model in Remote Sensing Images Based on Multi-Scale

 

IEEE Access SPECIAL SECTION ON DATA MINING FOR INTERNET OF THINGS

Received June 27, 2019, accepted July 8, 2019, date of publication July 15, 2019, date of current version July 31, 2019.

 

2. 요약

INDEX TERMS : Object detection, satellite imagery, pixel-level attention, spatial attention.

 

- multi-attention object detection method (MA-FPN)

- Multi-Attention Object Detection Model in Remote Sensing Images Based on Multi-Scale the network pay attention to the location of the object and reduce the loss of small object information

- According to feature pyramid network (FPN), we firstly put forward a global spatial attention module, which extracts spatial location-related information from shallow features and fuses it with deep features to enhance the position expression ability of deep features.

- pixel feature attention module: the multi-scale convolution kernel is employed to generate the feature map of the same size as the input