Abstract Birds are among the best bio-indicators, which can guide us to recognize some of the main conservation concerns in ecosystems. Anthropogenic impacts such as deforestation, habitat degradation, modification of landscapes, and decreased quality of habitats are major threats to bird diversity. The present study was designed to detect anthropogenic causative agents that act on waterbird diversity in Tarbella Dam, Indus River, Pakistan. Waterbird censuses were carried out from March 2019 to February 2020 in multiple areas around the dam. A total of 2990 waterbirds representing 63 species were recorded. We detected the highest waterbird richness and diversity at Pehure whereas the highest density was recorded at Kabbal. Human activity impacts seemed to be the main factor determining the waterbird communities as waterbirds were negatively correlated with the greatest anthropogenic impacts. Waterbirds seem to respond rapidly to human disturbance.
Human activity recognition is a key to a lot of applications such as healthcare and smart home. In this study, we provide a comprehensive survey on recent advances and challenges in human activity recognition (HAR) with deep learning. Although there are many surveys on HAR, they focused mainly on the taxonomy of HAR and reviewed the state-of-the-art HAR systems implemented with conventional machine learning methods. Recently, several works have also been done on reviewing studies that use deep models for HAR, whereas these works cover few deep models and their variants. There is still a need for a comprehensive and in-depth survey on HAR with recently developed deep learning methods.
This work aims to develop a novel fuzzy associator rule-based fuzzified deep convolutional neural network (FDCNN) architecture for the classification of smartphone sensor-based human activity recognition. This work mainly focuses on fusing the λmax method for weight initialization, as a data normalization technique, to achieve high accuracy of classification.
The major contributions of this work are modeled as FDCNN architecture, which is initially fused with a fuzzy logic based data aggregator. This work significantly focuses on normalizing the University of California, Irvine data set’s statistical parameters before feeding that to convolutional neural network layers. This FDCNN model with λmax method is instrumental in ensuring the faster convergence with improved performance accuracy in sensor based human activity recognition. Impact analysis is carried out to validate the appropriateness of the results with hyper-parameter tuning on the proposed FDCNN model with λmax method.
The effectiveness of the proposed FDCNN model with λmax method was outperformed than state-of-the-art models and attained with overall accuracy of 97.89% with overall F1 score as 0.9795.
The proposed fuzzy associate rule layer (FAL) layer is responsible for feature association based on fuzzy rules and regulates the uncertainty in the sensor data because of signal inferences and noises. Also, the normalized data is subjectively grouped based on the FAL kernel structure weights assigned with the λmax method.
Contributed a novel FDCNN architecture that can support those who are keen in advancing human activity recognition (HAR) recognition.
A novel FDCNN architecture is implemented with appropriate FAL kernel structures.
A changing climate makes the evaluation of human impacts on natural systems increasingly uncertain and affects the risk associated with management decisions. This influences both the achievability and meaning of marine conservation and resource management objectives. A risk-based framework that includes a risk equivalence approach in the evaluation of the potential consequences from human activity, can be a powerful tool for timely and consistent handling of environmental considerations in management advice. Risk equivalence permits a formal treatment of all sources of uncertainty, such that objectives-based management decisions can be maintained within acceptable risk levels and deliver outcomes consistent with expectations. There are two pathways to risk equivalence that can be used to account for the short-term and longer-term impacts of a changing environment: adjusting the degree of exposure to human pressure and adjusting the reference levels used to measure the risk. The first uses existing data and knowledge to derive risk conditioning factors applied to condition management advice on environmental departures from baseline conditions. The second is used to formalise the review and update of management objectives, reference levels and risk tolerances, so they remain consistent with potential consequences from human activity under new biological, ecological and socio-economic realities. A risk equivalence approach is about adapting existing practice to frame environmental considerations within objectives-based risk frameworks, systematically exploring alternative scenarios and assumptions, and conditioning management advice on environmental status. It is applicable to the management of all human activities impacting biological and ecological systems. Concepts of risk, risk conditioning factors, and incremental changes in risk, provide a common currency for the inclusion and communication of environmental effects into advice. Risk equivalence can ensure timely delivery of robust management advice accounting for demonstrated, anticipated or projected environmental effects. This can guide management decisions in a changing world, and greatly facilitate the implementation of an ecosystem approach for the management of human activities.
Due to the wide application of human activity recognition (HAR) in sports and health, a large number of HAR models based on deep learning have been proposed. However, many existing models ignore the effective extraction of spatial and temporal features of human activity data. This paper proposes a deep learning model based on residual block and bi-directional LSTM (BiLSTM). The model first extracts spatial features of multidimensional signals of MEMS inertial sensors automatically using the residual block, and then obtains the forward and backward dependencies of feature sequence using BiLSTM. Finally, the obtained features are fed into the Softmax layer to complete the human activity recognition. The optimal parameters of the model are obtained by experiments. A homemade dataset containing six common human activities of sitting, standing, walking, running, going upstairs and going downstairs is developed. The proposed model is evaluated on our dataset and two public datasets, WISDM and PAMAP2. The experimental results show that the proposed model achieves the accuracy of 96.95%, 97.32% and 97.15% on our dataset, WISDM and PAMAP2, respectively. Compared with some existing models, the proposed model has better performance and fewer parameters.
In humid environments, the formation of biofilms and microfouling are known to be the detrimental processes that first occur on stainless steel surfaces. This is known as biofouling. Subsequently, the conditions created by metabolites and the activity of organisms trigger corrosion of the metal and accelerate corrosion locally, causing a deterioration in, and alterations to, the performance of devices made of stainless steel. The microorganisms which thus affect stainless steel are mainly algae and bacteria. Within the macroorganisms that then damage the steel, mollusks and crustaceans are the most commonly observed. The aim of this review was to identify the mechanisms involved in biofouling on stainless steel and to evaluate the research done on preventing or mitigating this problem using nanotechnology in humid environments in three areas of human activity: food manufacturing, the implantation of medical devices, and infrastructure in marine settings. Of these protective processes that modify the steel surfaces, three approaches were examined: the use of inorganic nanoparticles; the use of polymeric coatings; and, finally, the generation of nanotextures.
A multiproxy (pollen, microcharcoal, loss-on-ignition, magnetic susceptibility and geochemistry) sequence from Lough Cullin, southeast Ireland, supported by a high-resolution radiocarbon chronology, modelled using Bayesian approaches, provides a record of environmental change for much of the Holocene. Following the establishment of mixed deciduous woodland, climatic deterioration was likely responsible for pronounced vegetation change and erosion, 7615–6500 cal. BC to 6245–5575 cal. BC, evidence for the ‘8.2 Kyr’ BP climate event. The so-called ‘elm decline’ is dated to 4220–3980 cal. BC and whilst there are possible indications of an anthropogenic cause, clear evidence of woodland clearance with cereal pollen is recorded at 3900–3700 cal. BC, 3790–3580 cal. BC and 3760–3650 cal. BC, during a period of clearance and farming of 320–450 years duration. A reduction in farming/settlement and woodland regeneration during the Middle Neolithic parallels the archaeological record, with low levels of activity during the Late Neolithic/Chalcolithic after 2960–2525 cal. BC, prior to increases during the Bronze Age then woodland clearance and agriculture between 1500–1410 and 1275–1000 cal. BC, corresponding with the archaeological evidence. A subsequent ‘step-wise’ reduction in human activity follows, from the latter date to 815–685 cal. BC, and a brief but pronounced cessation at 690–535 cal. BC. Renewed woodland clearance and agriculture commenced until 415–250 cal. BC. From the latter date until cal. AD 390–540, the Late Iron Age/Early Medieval period, a phase of woodland recovery is attested, followed by renewed landscape disturbance and arable agriculture in particular, continuing to the close of the record at cal. AD 780–1035.
AbstractHuman activity recognition (HAR) is a line of research whose goal is to design and develop automatic techniques for recognizing activities of daily living (ADLs) using signals from sensors. HAR is an active research filed in response to the ever-increasing need to collect information remotely related to ADLs for diagnostic and therapeutic purposes. Traditionally, HAR used environmental or wearable sensors to acquire signals and relied on traditional machine-learning techniques to classify ADLs. In recent years, HAR is moving towards the use of both wearable devices (such as smartphones or fitness trackers, since they are daily used by people and they include reliable inertial sensors), and deep learning techniques (given the encouraging results obtained in the area of computer vision). One of the major challenges related to HAR is population diversity, which makes difficult traditional machine-learning algorithms to generalize. Recently, researchers successfully attempted to address the problem by proposing techniques based on personalization combined with traditional machine learning. To date, no effort has been directed at investigating the benefits that personalization can bring in deep learning techniques in the HAR domain. The goal of our research is to verify if personalization applied to both traditional and deep learning techniques can lead to better performance than classical approaches (i.e., without personalization). The experiments were conducted on three datasets that are extensively used in the literature and that contain metadata related to the subjects. AdaBoost is the technique chosen for traditional machine learning, while convolutional neural network is the one chosen for deep learning. These techniques have shown to offer good performance. Personalization considers both the physical characteristics of the subjects and the inertial signals generated by the subjects. Results suggest that personalization is most effective when applied to traditional machine-learning techniques rather than to deep learning ones. Moreover, results show that deep learning without personalization performs better than any other methods experimented in the paper in those cases where the number of training samples is high and samples are heterogeneous (i.e., they represent a wider spectrum of the population). This suggests that traditional deep learning can be more effective, provided you have a large and heterogeneous dataset, intrinsically modeling the population diversity in the training process.