Abstract
Quantitative assessment of mitral valve (MV) morphology is important for diagnosing MV pathology and for planning of reparative procedures. Although this is typically done using 3D transesophageal echocardiography (TEE), recent advances in the spatiotemporal resolution of 3D transthoracic echocardiography (TTE) have enabled the use of this more patient friendly modality. However, manual data analysis is time consuming and operator dependent. In this study, a fully automatic method for MV segmentation and tracking in 3D TTE is proposed and validated. The proposed framework takes advantage of a previously proposed left ventricle (LV) segmentation framework to localize the MV and performs segmentation based on the B-spline Explicit Active Surfaces (BEAS) framework. The orientation of the MV is obtained and the MV surface is cropped to the mitral annulus (MA) and divided into posterior and anterior leaflets. The segmented MV at end diastole (ED) is propagated to end systole (ES) using localized anatomical affine optical flow (lAAOF). Because the orientation and leaflet division is known, relevant clinical parameters can then be extracted from the mesh at any time point. The proposed framework shows excellent segmentation results with a mean absolute distance (MAD) and Hausdorff distance (HD) of 1.19 +/- 0.25 mm and 5.79 +/- 1.25 mm at ED and 1.39 +/- 0.32 mm and 6.70 +/- 1.97 mm at ES against manual analysis. In conclusion, an automatic method for MV segmentation is proposed which could provide valuable clinical information in a more patient-friendly manner.

Abstract
Panoramic or aerial images can be acquired with some easiness and cover vast tracts of territory to be used in fire detection. The analysis of these images, in particular based on color and threshold indices, can be very interesting computationally when applied in real time systems and collected, for example, through drones or watchtowers. This paper presents a solution designated Color Algorithm for Flame Exposure (CAFE), which significantly improves an existing method (cf. Forest Fire Detection Index - FFDI) in flame detection, based on daylight images, in mixed Mediterranean landscape, containing vegetation, buildings, burning areas, land, etc. The CAFE approach, presented, adds a parameterizable transformation of the image into the Lab color space. This approach was tested in four distinct scenarios, significantly reducing false positives and maintaining an equivalent level of false negatives when compared to the FFDI approach.

Abstract

Abstract
Securing wireless communication is significant for privacy and confidentiality of sensing data in Cyber-Physical Systems (CPS). However, due to broadcast nature of radio channels, disseminating sensory data is vulnerable to eavesdropping and message modification. Generating secret keys by extracting the shared randomness in a wireless fading channel is a promising way to improve the communication security. In this poster, we present a novel secret key generation protocol for securing real-time data dissemination in CPS, where the sensor nodes cooperatively generate a shared key by estimating the quantized fading channel randomness. A 2-hop wireless sensor network testbed is built and preliminary experimental results show that the quantization intervals and distance between the nodes lead to a secret bit mismatch.

Abstract
Data networks are naturally prone to interferences that can corrupt messages, leading to performance degradation or even to critical failure of the corresponding distributed system. To improve resilience of critical systems, time-triggered networks are frequently used, based on communication schedules defined at design-time. These networks offer prompt error detection, but slow error recovery that can only be compensated with bandwidth overprovisioning. On the contrary, the Flexible Time-Triggered (FTT) paradigm uses online traffic scheduling, which enables a compromise between error detection and recovery that can achieve timely recovery with a fraction of the needed bandwidth. This article presents a new method to recover transmission errors in a time-triggered Controller Area Network (CAN) network, based on the Flexible Time-Triggered paradigm, namely FTT-CAN. The method is based on using a server (traffic shaper) to regulate the retransmission of corrupted or omitted messages. We show how to design the server to simultaneously: (1) meet a predefined reliability goal, when considering worst case error recovery scenarios bounded probabilistically by a Poisson process that models the fault arrival rate; and, (2) limit the direct and indirect interference in the message set, preserving overall system schedulability. Extensive simulations with multiple scenarios, based on practical and randomly generated systems, show a reduction of two orders of magnitude in the average bandwidth taken by the proposed error recovery mechanism, when compared with traditional approaches available in the literature based on adding extra pre-defined transmission slots.

Abstract
Precision agriculture is gaining an increasing interest in the current farming paradigm. This new production concept relies on the use of information technology (IT) to provide a control and supervising structure that can lead to better management policies. In this framework, imaging techniques that provide visual information over the farming area play an important role in production status monitoring. As such, accurate representation of the gathered production images is amajor concern, especially if those images are used in detection and classification tasks. Real scenes, observed in natural environment, present high dynamic ranges that cannot be represented by the common LDR (Low Dynamic Range) devices. However, this issue can be handled by High Dynamic Range (HDR) images since they have the ability to store luminance information similarly to the human visual system. In order to prove their advantage in image processing, a comparative analysis between LDR and HDR images, for fruits detection and counting, was carried out. The obtained results show that the use of HDR images improves the detection performance to more than 30% when compared to LDR.