In 2021, compared to the last year, global port container logistic volumes increased by 7% (857 million 20-foot equivalent units (TEU)), while deliveries of newly built container ships increased by 18% (10,929 thousand gross tons) [1]. Hence, the problem instigated by an increased volume of logistics and port congestion can be solved through the efficient operation of the port, and ship detection and volume data processing have become important. Ship monitoring related to port logistics is possible by checking the Auto Identification System (AIS) [2]. The Terrestrial AIS (T-AIS) system is capable of detecting only short-range communication up to 50 nautical miles [3], and there is an omission to monitor the ship when the navigator controls the transmission-receiving of AIS [4]. Therefore, as the ability of satellites increases, there is a movement to detect missing ships using satellite images [5]. In this study, ship detection was conducted to monitor container ships that were not operate in the anchorage area around the port or berth at the pier. In order to find container ships through deep learning, four types of vessels (container ships, bulk ships, oil tankers, and small ships) were extracted from Compact Advanced Satellite 500-1 (CAS500-1) satellite images in the yellow sea and one port to make a training dataset. The CAS500 satellites carry different sensors for different Earth observation missions. CAS500-1 has the panchromatic and multispectral modes using the AEISS-C (Advanced Earth Imaging Sensor System) payload for land surface imagery, land topography imagery, and vegetation-type imagery. Although improvement in remote surface reflectance is needed, it can be beneficial at sea. The detection model was established through YOLOv5 (You Only Look Once version 5) [6] deep learning algorithm. The model's performance was evaluated by applying them to the CAS500-1, and Korea Multi-Purpose Satellite-3 (KOMPSAT-3), taken at the Busan New Port, Korea, and Oakland Port in the U.S. The validation was performed based on mated YOLO results and AIS. In this process, the ship type, which is static information of the ship, and the position value, which is dynamic information, were compared. Ship detection results presented an accuracy of 66% to 93%, and the classification results had an accuracy of 70% to 89%. In particular, the container ships interested in this study were accurately classified as container ships of at least 85% up to 100%. This study implies that it attempts to classify vessel types in satellite images using deep learning and evaluates the possibility of ship detection and ship classification for types based on the CAS500-1 satellite image. This study concludes that if a deep learning model is applied to the CAS500-1 and multi satellite to detect the ship's coordinate and type, it can contribute to the management of shipping traffic and monitoring of ship traffic for autonomous driving ships. Also, the results indicate that high resolution satellite images can be used in mooring ships for vessel monitoring. The developed approach can potentially be used in vessel tracking and monitoring systems at major ports around the world if the accuracy of the detection model is improved through continuous learning data construction.

MOTIVATION: Cellular behavior is determined by complex non-linear interactions between numerous intracellular molecules that are often represented by Boolean network models. To achieve a desired cellular behavior with minimal intervention, we need to identify optimal control targets that can drive heterogeneous cellular states to the desired phenotypic cellular state with minimal node intervention. Previous attempts to realize such global stabilization were based solely on either network structure information or simple linear dynamics. Other attempts based on non-linear dynamics are not scalable. RESULTS: Here, we investigate the underlying relationship between structurally identified control targets and optimal global stabilizing control targets based on non-linear dynamics. We discovered that optimal global stabilizing control targets can be identified by analyzing the dynamics between structurally identified control targets. Utilizing these findings, we developed a scalable global stabilizing control framework using both structural and dynamic information. Our framework narrows down the search space based on strongly connected components and feedback vertex sets then identifies global stabilizing control targets based on the canalization of Boolean network dynamics. We find that the proposed global stabilizing control is superior with respect to the number of control target nodes, scalability, and computational complexity. AVAILABILITY AND IMPLEMENTATION: We provide a GitHub repository that contains the DCGS framework written in Python as well as biological random Boolean network datasets (https://github.com/sugyun/DCGS). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

In this study, we computed the Radar Vegetation Index (RVI) using observations made with a ground based multi-frequency polarimetric scatterometer system over an entire wheat growth period. The temporal variations of the backscattering coefficients for L-, C-, and X-band, RVI, Vegetation water content (VWC), and fresh weight were analyzed. We found that the L-band RVI was very sensitive to VWC and fresh weight. Based on the correlation analysis between RVI and these growth parameters, we developed equations for estimation VWC and fresh weight.