D3.1: Thermal/infrared sensing

Posted: 20th Oct 2016  -  Download

The SUNNY project aims to develop and integrate a new tool for collecting real-time information using heterogeneous sensors carried by multiple UAV platforms, and for analyzing this information automatically to provide situational awareness for the system operators to monitor the EU maritime borders more effectively and efficiently. In this document, we will describe the specification, performance and qualification tests of the infrared sensors. Two infrared sensors are developed by Xenics: a SWIR sensor (XSW) and a LWIR sensor (XTM).

WP1.4 Surveillance Societal and Ethetical Aspects

Posted: 10th Jan 2015  -  Download

This document details issues that need to be taken into account with designing SUNNY capabilities: a. Data protection and privacy issues b. Privacy by Design (PbD) and the impact of the concept in SUNNY c. Current EU border management with examples of irregular migration in specific regions. d. Legal and human rights issues

D3.3: Radar sensing

Posted: 24th Oct 2016  -  Download

In this document the main customizations of the MetaSensing’s MiniSAR sensor are summarized, which have been introduced to match with the SUNNY specifications; the processing related to the generation of ISAR images from radar data is overviewed, and the equipment which will be actually used during the SUNNY demonstrator are listed; the main results of the qualification tests will be finally shown, assessing the expected performance of the radar system.

D3.4: Sensor Alignment and Fusion

Posted: 24th Oct 2016  -  Download

The objective of this deliverable D3.4 is to describe the sensor alignment and fusion module developed in the SUNNY project. The focus is on development of advanced data fusion algorithms for effectively combining information collected from multiple sensors each of which could be noisy but possibly complementary to each other. In particular, different types of sensory data will have different properties and capture distinct aspects and information of the same physical scene. Algorithms are developed to align the data from different sensors and improve the image spatial resolution for facilitating information fusion and decision making. These algorithms will be validated with the SUNNY datasets and in the final demonstration.

D4.1 On-board data processing

Posted: 24th Oct 2016  -  Download

This deliverable addresses the development of an integrated solution for UAV on-board data processing. This report describes the SUNNY on-board data processing system based on the specific functional sub­system architecture and requirements for the data acquisition of on-board data processing of the SUNNY system. The deliverable addresses the following topics: some related work on each individual sensor processing, and known subsystem modules. It addresses how the SUNNY on-board data processing complies with the requirements imposed by the users and also by the global SUNNY system architecture. Based on these requirements, we design a on-board data processing architecture capable of coping with the SUNNY system objectives, and specify each individual module. Individual sensor integration and performance is analyzed, and the data flow between on-board and ground system is defined. Results of the real-time image processing of electro/optical cameras, infrared cameras and hyperspectral cameras are also presented. This deliverable also contains a section detailing the dataset campaigns, including flights performed and data collection payload used.

D4.2: Automated Target Identification

Posted: 24th Oct 2016  -  Download

The objective of D4.2 is to describe the automated target identification module developed in the SUNNY project. This report starts by introducing the SUNNY project context (Chapter 1), based on which the requirements for the target identification and active learning modules are given (Chapter 2). We then review the state-of-the-art target detection models and algorithms (Chapter 3). In Chapter 4, we first present the details of our target identification algorithms using single sensor, and then describe how multiple sensor data can be effectively combined to improve target identification performance. These algorithms are evaluated on a number of multi-sensor SUNNY datasets. In Chapter 5, we detail the active learning algorithm with the aim to assist the target identification model learning with less manual annotation cost. Lastly, we describe the software implementation details and the interaction of the two modules with other sibling modules in the entire SUNNY system (Chapter 6). The deliverable concludes in Chapter 7.