A5046318445- Planetary Science and Exploration
- Remote-Sensing Image Classification
- Remote Sensing and Land Use
- Astro and Planetary Science
- Remote Sensing in Agriculture
- Laser-induced spectroscopy and plasma
- Geochemistry and Geologic Mapping
- Spaceflight effects on biology
- Space Science and Extraterrestrial Life
- Advanced Image and Video Retrieval Techniques
- Advanced Chemical Sensor Technologies
- Geology and Paleoclimatology Research
- Isotope Analysis in Ecology
- Space exploration and regulation
- Robotics and Sensor-Based Localization
- Spectroscopy and Chemometric Analyses
- Advanced Image Fusion Techniques
Tongji University
2021-2025
Collaborative Innovation Centre for Advanced Ship and Deep-Sea Exploration
2025
Multi-temporal change detection (CD) plays a crucial role in the remote sensing application field. In recent years, supervised deep learning methods have shown excellent performance detecting changes very-high-resolution (VHR) images. However, these require large number of labeled samples for training, making process time-consuming and labor-intensive. Unsupervised approaches are more attractive practical applications since they can produce CD map without relying on any ground reference or...
On May 15, 2021, China's first Mars rover, the Zhurong rover successfully landed on Utopian Planitia in northern region of Mars. The multispectral camera (MSCam) board has captured multi-spectral images, which provide spatial and spectral information about in-situ observation targets facilitate analysis types materials Martian surface. However, frequent sandstorms are accompanied by dust deposition, varying degrees coverage have altered original characteristics scientific detection targets,...
Landform classification and mapping of the Martian surface using Mars orbiter images can provide an important reference for landing site selection rovers' traversability evaluation in exploration. Moreover, specific landforms are closely associated with evidences water-related activities life, thus have crucial research importance. This article proposes a novel superpixel-guided multiview feature fusion network (MarsMapNet) efficient landforms. In particular, proposed MarsMapNet first...
Among various multimodal remote sensing data, the pairing of multispectral (MS) and panchromatic (PAN) images is widely used in applications. This article proposes a novel global collaborative fusion network (GCFnet) for joint classification MS PAN images. In particular, patch-free scheme based on an encoder-decoder deep learning (DL) developed to exploit context dependencies image. The proposed GCFnet designed architecture, which mainly contains three parts: 1) two shallow-to-deep feature...
The Tianwen-1 mission, China's first interplanetary endeavor and Mars Mission, touched the surface of Red Planet on May 15, 2021. With successful landing Zhurong rover southern Utopian Plain Mars, Surface Composition Detector (MarSCoDe) board has started to analyze material composition Martian by using Laser Induced Breakdown Spectroscopy (LIBS). However, changes in instrument temperature external environment during operation will cause spectral drift LIBS data, leading inaccurate inversion...
多时相遥感影像变化检测是指对同一地理区域、不同时间获取的遥感影像进行自动变化发现、识别与解释的遥感处理与分析技术。随着卫星遥感技术及人工智能理论方法的快速发展,基于多时相遥感影像数据驱动和模型驱动的传统变化检测方法正朝着数据—模型—知识联合驱动的方向转型和演变,以更加自动化、精细化和智能化的方式,解决多领域的地表时空变化检测问题。本文在总结多时相遥感数据源从同构到异构、变化检测模型从传统到智能、变化检测应用从理论到落地过程中存在问题的基础上,以光学遥感影像变化检测任务为例,梳理和分析了人工智能时代下变化检测技术的发展历程。从无监督、监督、弱监督3个方面探讨了遥感变化检测从传统到前沿技术的转型特点与趋势,并进一步提出了未来需重点突破模型的物理可解释性、泛化及迁移能力、跨数据—跨场景—跨领域应用水平等关键问题。
Hyperspectral images (HSIs) provides abundant spectral information through hundreds of bands with continuous that can be used in land cover fine change detection (CD). HSIs make it possible for hyperspectral CD performance higher discrimination on changes but a challenge to the conventional techniques due its high dimensionality and dense representation. In this paper, we implemented intrinsic image decomposition (IID) model decompose temporal difference into two parts: real pseudo...
Comparing with the multispectral remote sensing image, hyperspectral image (HSI) has higher spectral resolution, a near continuous signature, thus can represent fine variations that occurred in temporal domain. This allows more changes to be detected, especially major reflected on overall signature (associating abrupt land-cover transitions), as well subtle reflect only portion of change physicochemical properties classes). Currently, there are some available detection (CD) data sets....