Abstract
Deep transfer learning emerged as a new paradigm in machine learning in which a deep model is trained on a source task and the knowledge acquired is then totally or partially transferred to help in solving a target task. In this paper, we apply the source-target-source methodology, both in its original form and an extended multi-source version, to the problem of cross-sensor biometric recognition. We tested the proposed methodology on the publicly available CSIP image database, achieving state-of-the-art results in a wide variety of cross-sensor scenarios.

Abstract
The rising challenges in the fields of iris and face recognition are leading to a renewed interest in the area. In recent years the focus of research has turned towards alternative traits to aid in the recognition process under less constrained image acquisition conditions. The present work assesses the potential of the periocular region as an alternative to both iris and face in such scenarios. An automatic modeling of SIFT descriptors, regardless of the number of detected keypoints and using a GMM-based Universal Background Model method, is proposed. This framework is based on the Universal Background Model strategy, first proposed for speaker verification, extrapolated into an image-based application. Such approach allows a tight coupling between individual models and a robust likelihood-ratio decision step. The algorithm was tested on the UBIRIS.v2 and the MobBIO databases and presented state-of-the-art performance for a variety of experimental setups.

Abstract
The duration of 802.11 user sessions has been widely studied in the context of analyzing user behavior and mobility. Short (smaller-than-5-minutes) sessions are never used or characterized in these analyses as they are unrelated to user behavior and considered as artifacts introduced by the wireless network. In this paper we characterize short 802.11 sessions as recorded through RADIUS authentication. We show that 50% of access points have 70% of smaller than 5 minutes sessions in a 5 months trace from the Eduroam academic wireless network in the University of Porto. Exactly because they are artifacts introduced by the network, short sessions are an important indicator for network management and the quality of the wireless access. Network managers typically do not collect and process session information but rely on SNMP to provide summaries of 802.11 usage data. We develop a modeling framework to provide predictions for the number of short sessions from SNMP data. We model the data stream of each access point using two methods of regression and one classification technique. We evaluate these models based on short session prediction accuracy. The models are trained on the 5 months data and the best results show prediction accuracy of 95.27% in polynomial regression at degree of 3.

Abstract
"Taking less treating better" has been one of the major improvements of breast cancer surgery in the last four decades. The application of this principle translates into equivalent survival of breast cancer conserving treatment (BCT) when compared to mastectomy, with a better cosmetic outcome. While it is relatively easy to evaluate the oncological results of BCT, the cosmetic outcome is more difficult to measure due to the lack of an effective and consensual procedure. The assessment of cosmetic outcome has been mainly subjective, undertaken by a panel of expert observers or/and by patient self-assessment. Unfortunately, the reproducibility of these methods is low. Objective methods have higher values of reproducibility but still lack the inclusion of several features considered by specialists in BCT to be fundamental for cosmetic outcome. The recent addition of volume information obtained with 3D images seems promising. Until now, unfortunately, no method is considered to be the standard of care. This paper revises the history of cosmetic evaluation and guides us into the future aiming at a method that can easily be used and accepted by all, caregivers and caretakers, allowing not only the comparison of results but the improvement of performance. (C) 2014 Wiley Periodicals, Inc.

Abstract
In this paper, a new method for music symbol classification named Combined Neural Network (CNN) is proposed. Tests are conducted on more than 9000 music symbols from both real and scanned music sheets, which show that the proposed technique offers superior classification capability. At the same time, the performance of the new network is compared with the single Neural Network (NN) classifier using the same music scores. The average classification accuracy increased more than ten percent, reaching 98.82%.

Abstract
Animal behavior assessment plays an important role in basic and clinical neuroscience. Although assessing the higher functional level of the nervous system is already possible, behavioral tests are extremely complex to design and analyze. Animal's responses are often evaluated manually, making it subjective, extremely time consuming, poorly reproducible and potentially fallible. The main goal of the present work is to evaluate the use of consumer depth cameras, such as the Microsoft's Kinect, for detection of behavioral patterns of mice. The hypothesis is that the depth information, should enable a more feasible and robust method for automatic behavior recognition. Thus, we introduce our depth-map based approach comprising mouse segmentation, body-like per-frame feature extraction and per-frame classification given temporal context, to prove the usability of this methodology.