Cropland Monitoring Based on Remote Sensing Imagery

Dear Colleagues,

Remote sensing has long been used in monitoring agricultural activities, including crop type mapping, yield prediction, crop phenology, and crop management. During the past years, key trends in crop monitoring using remote sensing evolved over time, among a few examples:

• Efforts have been devoted to model generalization. While many approaches have been successfully proposed for monitoring crop growth, it is often challenging to apply the model to a wider spatial and temporal domain without recalibration. The lack of automation has been a major obstacle that hinders large scale agricultural applications. To address the issue, studies have been carried out to enhance model transferability by incorporating intrinsic physical characteristics and processes of crops into the monitoring approach. With the aid of process-based crop models and data assimilation, the interaction between weather, soil and water factors as well as the crop-specific responses to a range of environmental dynamics can be represented. Physics-guided deep learning offers a unique opportunity to build a high-performance, explainable, and generalizable modelling framework. These emerging techniques are especially promising for real-time observations and models to monitor crops during an early stage without the need of time-consuming re-training processes. At present it is still difficult to build highly transferable relationships across distinct landscapes.
• Recent advances in deep learning have provided unprecedentedly effective means to model complex spatial patterns and temporal dependencies. Deep neural networks are able to achieve higher accuracies than ever in the identification of cropland texture and special land features, making it possible to track cropland surface conditions precisely and discover agricultural practices such as the use of pivot irrigation and silo bags. Time series analysis is the core of many cropland monitoring tasks, for which deep learning-based approaches extract short-term and long-term data connections simultaneously and build universal feature representations for temporal trajectories of any shape. Irregular and highly variant temporal patterns are hard to model by conventional approaches, for example, cover crops and silage crops may benefit from deep neural networks and big data. Undoubtedly deep learning has been a particularly active research field in cropland monitoring.
• Researchers and agricultural practitioners now have growing access to new sensors and instruments like UAV, LiDAR, and flux towers. With improved data quality and platform capability, it is also becoming easier to use the data synergically across scales. The new multi-source and multi-scale remote sensing data sources offer great potential for crop monitoring by providing complementary information in spatial, temporal, structural or spectral dimensions to overcome the challenges in conventional practices. However, more efforts are still needed to develop effective frameworks to fully utilize the multi-source data and achieve the goal of more accurate and efficient crop monitoring.

The proposed special issue will distribute studies of the recent development in crop monitoring to a broader audience. Articles covering but not limited to the aforementioned topics are cordially invited to this special issue.

Dr. Liheng Zhong
Prof. Dr. Ran Meng
Prof. Dr. Ignacio A. Ciampitti
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• Cropland monitoring
• Crop model
• Simulation model
• Biophysical modelling
• Deep learning
• Data assimilation
• Time series analysis
• Data fusion
• Multi-source remote sensing
• Satellite imagery
• UAV