# A Comparison and Strategy of Semantic Segmentation on Remote Sensing Images

## TL;DR

This paper compares recent deep learning models for satellite image semantic segmentation, analyzing their accuracy and resource use, and proposes a strategy suitable for on-orbit deployment on satellites like TianZhi-2.

## Contribution

It provides a comprehensive comparison of models considering satellite constraints and introduces a practical on-orbit segmentation strategy for deployment.

## Key findings

- Deep learning models vary in accuracy and resource consumption.
- A viable on-orbit segmentation strategy is proposed.
- Strategy will be deployed on TianZhi-2 satellite.

## Abstract

In recent years, with the development of aerospace technology, we use more and more images captured by satellites to obtain information. But a large number of useless raw images, limited data storage resource and poor transmission capability on satellites hinder our use of valuable images. Therefore, it is necessary to deploy an on-orbit semantic segmentation model to filter out useless images before data transmission. In this paper, we present a detailed comparison on the recent deep learning models. Considering the computing environment of satellites, we compare methods from accuracy, parameters and resource consumption on the same public dataset. And we also analyze the relation between them. Based on experimental results, we further propose a viable on-orbit semantic segmentation strategy. It will be deployed on the TianZhi-2 satellite which supports deep learning methods and will be lunched soon.