中国吉林一号太空遥感卫星China Jilin-1 space remote sensing satellite

“吉林一号”卫星是中国长光卫星技术有限公司在建的核心工程，是中国重要的太空光学遥感卫星。预计2025年左右，“吉林一号”将实现138颗卫星组网。卫星建设分为两个阶段：第一阶段实现60颗卫星在轨组网，具备全球热点地区30分钟内重访能力，每天可观测全世界范围内800多个目标区域；第二阶段实现138颗卫星在轨组网，具备全球任意地点10分钟内重访能力。卫星覆盖面积累计达1.33亿平方公里，具备全球一张图一年更新1次、全中国一张图一年更新3次的能力。依托“吉林一号”遥感数据，中国长光卫星利用人工智能算法，实现了全中国地表分类的精准化、自动化更新，可以在线完成建筑、道路、水体、林木以及不同类型农作物的自动识别，识别精度超过95%。具有响应速度快、覆盖面积广的特点。

前面我们也提到了，“吉林一号”是由多颗卫星组成的星座，它不仅可以在移动的过程中自动调整镜头方向，锁住固定甚至移动的目标，然后实时拍照或录像并传回地面，而且还可以在前面的卫星走了以后，后面的卫星接力追踪，另外，它还可以利用AI技术对目标进行自动识别，无需人工标定，这样侦察效率就会非常高。这项功能有点类似于能够识别、跟踪和锁定目标的火控雷达。
The "Jilin-1" satellite is a core project under construction by China Changguang Satellite Technology Co., Ltd. and an important space optical remote sensing satellite in China. It is expected that around 2025, "Jilin-1" will realize a network of 138 satellites. The satellite construction is divided into two stages: in the first stage, 60 satellites are in orbit, capable of revisiting global hotspots within 30 minutes, and can observe more than 800 target areas around the world every day; in the second stage, 138 satellites are realized. The satellite is in orbit and has the ability to revisit anywhere in the world within 10 minutes. The satellite covers a cumulative area of ​​133 million square kilometers, with the ability to update a map of the world once a year and a map of China three times a year. Relying on the "Jilin No. 1" remote sensing data, China Changguang Satellite uses artificial intelligence algorithms to achieve accurate and automatic update of land surface classification across China, and can complete online automatic identification of buildings, roads, water bodies, trees and different types of crops. The accuracy is over 95%. It has the characteristics of fast response and wide coverage.

We also mentioned earlier that "Jilin-1" is a constellation composed of multiple satellites. It can not only automatically adjust the direction of the lens during the movement, lock fixed or even moving targets, and then take pictures or videos in real time and send them back to the ground. , and it can also track the satellites behind after the satellites in front have left. In addition, it can also use AI technology to automatically identify the target without manual calibration, so the reconnaissance efficiency will be very high. This function is somewhat similar to a fire control radar that can identify, track and lock onto targets.