Online self-supervised multi-instance segmentation of dynamic objects

This paper presents a method for the continuous segmentation of dynamic objects using only a vehicle mounted monocular camera without any prior knowledge of the object’s appearance. Prior work in online static/dynamic segmentation is extended to identify multiple instances of dynamic objects by introducing an unsupervised motion clustering step. These clusters are then used to update a multi-class classifier within a self-supervised framework. In contrast to many tracking-by-detection based methods, our system is able to detect dynamic objects without any prior knowledge of their visual appearance shape or location. Furthermore, the classifier is used to propagate labels of the same object in previous frames, which facilitates the continuous tracking of individual objects based on motion. The proposed system is evaluated using recall and false alarm metrics in addition to a new multi-instance labelled dataset to evaluate the performance of segmenting multiple instances of objects.

Partial mediation role of self-efficacy between positive social interaction and mental health in family caregivers for dementia patients in Shanghai

We explored the mediation effect of caregiver self-efficacy on the influences of behavioral and psychological symptoms (BPSD) of dementia care recipients (CRs) or family caregivers’ (CGs) social supports (informational, tangible and affectionate support and positive social interaction) on CGs’ mental health. We interviewed 196 CGs, using a battery of measures including demographic data of the dyads, CRs’ dementia-related impairments, and CGs’ social support, self-efficacy and the Medical Outcome Study (MOS) Short-Form (SF-36) Health Survey. Multiple regression analyses showed that gathering information on self-efficacy and managing CG distress self-efficacy were the partial mediators of the relationship between positive social interaction and CG mental health. Managing caregiving distress self-efficacy also partial mediated the impact of BPSD on CG mental health. We discuss implications of the results for improving mental health of the target population in mainland China.

Affordance in interaction

The concept of affordance has different interpretations in the field of Human-Computer Interaction (HCI). However, its treatment has been merely as a one-to-one relationship between a user and a technology. We believe that a broader view of affordances is needed which encompasses social and cultural aspects of our everyday life. We propose an interaction-centered view of affordance that can be useful for developing better understandings of designed artefacts. An interaction-centered view of affordance suggests that affordance is an interpretative relationship between users and the technology that emerges during the users' interaction with the technology in the lived environments. We distinguish two broad classes of affordances: affordance in Information and affordance in Articulation. Affordance in information refers to users' understanding of a technology based on their semantic and syntactic interpretation; and affordance in articulation refers to users' interpretations about the use of the technology. We also argue that the notion of affordance should be treated at two levels: at the 'artefact level' and at the 'practice level'. Consequently, we provide two examples to demonstrate our arguments.

Quantitative genetic parameters for body traits at different ages in a cultured stock of giant freshwater prawn (Macrobrachium rosenbergii) selected for fast growth

The aim of the current study was to estimate heritabilities and correlations for body traits at different ages (Weeks 10 and 18 after stocking) in a giant freshwater prawn (Macrobrachium rosenbergii) population selected for fast growth rate in Vietnam. The dataset consisted of 4650 body records (2432 and 2218 records collected at Weeks 10 and 18, respectively) in the full pedigree comprising a total of 18 387 records. Variance and covariance components were estimated using restricted maximum likelihood fitting a multi-trait animal model. Estimates of heritability for body traits (bodyweight, body length, cephalothorax length, abdominal length, cephalothorax width and abdominal width) were moderate and ranged from 0.06 to 0.11 and from 0.11 to 0.22 at Weeks 10 and 18, respectively. Body-trait heritabilities estimated at Week 10 were not significantly lower than at Week 18. Genetic correlations between body traits within age and genetic correlations for body traits between ages were generally high. Our results suggested that selection for high growth rate in GFP can be undertaken successfully before full market size has been reached.

Binary image steganographic techniques classification based on multi-class steganalysis

In this paper, we propose a new multi-class steganalysis for binary image. The proposed method can identify the type of steganographic technique used by examining on the given binary image. In addition, our proposed method is also capable of differentiating an image with hidden message from the one without hidden message. In order to do that, we will extract some features from the binary image. The feature extraction method used is a combination of the method extended from our previous work and some new methods proposed in this paper. Based on the extracted feature sets, we construct our multi-class steganalysis from the SVM classifier. We also present the empirical works to demonstrate that the proposed method can effectively identify five different types of steganography.

Health monitoring of short-and medium-spen bridges using an improved modal strain energy method

Increasing the importance and use of infrastructures such as bridges, demands more effective structural health monitoring (SHM) systems. SHM has well addressed the damage detection issues through several methods such as modal strain energy (MSE). Many of the available MSE methods either have been validated for limited type of structures such as beams or their performance is not satisfactory. Therefore, it requires a further improvement and validation of them for different types of structures. In this study, an MSE method was mathematically improved to precisely quantify the structural damage at an early stage of formation. Initially, the MSE equation was accurately formulated considering the damaged stiffness and then it was used for derivation of a more accurate sensitivity matrix. Verification of the improved method was done through two plane structures: a steel truss bridge and a concrete frame bridge models that demonstrate the framework of a short- and medium-span of bridge samples. Two damage scenarios including single- and multiple-damage were considered to occur in each structure. Then, for each structure, both intact and damaged, modal analysis was performed using STRAND7. Effects of up to 5 per cent noise were also comprised. The simulated mode shapes and natural frequencies derived were then imported to a MATLAB code. The results indicate that the improved method converges fast and performs well in agreement with numerical assumptions with few computational cycles. In presence of some noise level, it performs quite well too. The findings of this study can be numerically extended to 2D infrastructures particularly short- and medium-span bridges to detect the damage and quantify it more accurately. The method is capable of providing a proper SHM that facilitates timely maintenance of bridges to minimise the possible loss of lives and properties.