Event Streams Research Articles

Flow pulses shape periphyton differently according to local light and nutrient conditions in experimental lowland streams

Abstract Streams generally are affected by multiple stressors acting at different timescales. Periphyton, often the most important primary producer in these ecosystems, may respond to short‐term impacts as well as to different long‐term environmental conditions with potentially various changes in community structure. Here, we experimentally investigated the effects of sudden flow pulses on periphyton communities as a way to mimic extreme precipitation events in lowland streams that are predicted to occur more often with climate change in some regions. Using outdoor flumes, we allowed periphyton to colonise nutrient‐diffusion substrates under two light conditions (50% shade and fully open) and nutrient availabilities (control, with access only to stream nutrients, and N–P‐enriched) along a gradient in baseflows (0.43 to 2.17 L/s). After one month, we exposed the communities to a flow pulse (two‐fold peak flow increase to simulate conditions of a potentially high disturbance) and analysed the responses of biomass and taxonomic composition. Flow pulse promoted periphyton growth in the lowest range of the baseflow but led to biomass reduction in the highest range. Light was the second major driver of biomass accrual, whereas nutrient enrichment had a strong effect on community composition both before and after the pulse (i.e., diatom dominance vs. green algae dominance in scenarios without and with enrichment, respectively). In all treatments, the flow pulse promoted a higher taxonomic richness, suggesting a partial reset of the succession of the periphyton communities. However, independent of flow and resources, periphyton communities showed low ecological resistance against the pulse with changes in chlorophyll a, biovolume and taxonomic richness to the pulse. We demonstrated that the effects of pulses on periphyton are similar in terms of biomass but varied strongly regarding composition depending on their initial structure, which is in turn mediated by the baseflow normally experienced by the systems, and on light and nutrient availability. Our results highlight the importance of testing multiple stressors, such as an increase in extreme events, under a wide range of environmental conditions (i.e., flow regime, trophic state, light availability). This approach allows us to detect potential interaction effects and non‐linear responses, and highlights that the environmental settings ultimately determine the net effects of flow pulses on community structure and probably also on several important ecosystem processes.

PARAMETRIC OPTIMIZATION OF INFORMATION SYSTEMS FOR PROACTIVE PROTECTION OF DATA TRANSMISSION SERVICE FROM NETWORK RECONNAISSANCE

The analysis of the FTP protocol revealed a number of vulnerabilities in it, which can be used by an adversary to uncover the IDs of data transfer service clients and compromise the used network security features. On the one hand, it is due to the possibility of repeated authorization attempts to unauthorized clients, allowing to implement a data transfer service username and password. On the other hand, the implementation of reactive measures of protection, which consists in breaking the control connection to the adversary after a number of unsuccessful authorization attempts, forcing him to change the strategy of exposure. In this regard, the use of proactive protection measures, allowing to increase the time of dialog interaction with the intruder and imposing restrictions on computing and temporal resources used by the intruder, is proposed as a countermeasure to the network reconnaissance tools. The essence of proactive protection measures consists in splitting the response of FTP server to unauthorized client commands into fragments and sending those fragments to the intruder after a specified delay time. Application of proactive protection gives the security system of information system a number of advantages over the intruder in using the time resource, allowing to implement additional protection measures. The problematic issues when implementing proactive protection measures for data transmission service of information systems are, firstly, the need to send fragments of response to the intruder at intervals shorter than the timeout time of waiting for these responses, set by the intruder. Secondly, the possibility of sending to the violator the limited number of fragments of response, that is caused by FTP protocol capabilities, regulating the size of response, as well as the excessive saturation of the communication channel by fragments of messages, that can lead to exhaustion of the resource opportunities of data transmission medium. Therefore, there is a need to find the optimal parameters of the information system in the configuration of data transmission parameters in the interaction with the network reconnaissance tools. The process of information system՚s data transmission service functioning under the influence of network reconnaissance tools is formalized in the form of a Markov random process with discrete states and continuous time. Intensities of streams of events, under which the system passes from state to state, are the input data for solving the task of finding the optimal parameters of the information system for the situation under consideration. The qualitative formulation of the task of determining the optimal parameters of the information system is formulated as follows: using a given mathematical model of the information system՚s data transmission service under conditions of network reconnaissance we should find such a set of parameters of information system՚s functioning, in which the maximum efficiency of its protection while ensuring a given level of availability is provided. The problem under consideration is a vector (multi-criteria) parametric optimization problem, which will be reduced to single-criteria (scalar) optimization by constraint method (main criterion). Main criterion method implies selection of one scalar target function and transfer of other target functions into additional restriction functions. As the constraints for solving the optimization problem set resource constraints, characterized by the coefficient of the transmission medium, as well as the condition of a given level of robustness of the information system, characterized by the number of conditionality of the matrix of the intensity of the flow of events. The research applies a nonlinear optimization algorithm, an annealing simulation algorithm that provides the ability to approximate the search for a global extremum. The results obtained during the study are presented in the form of an assortment of graphic dependences of changes in the optimal parameters of the information system on the intensity of the impact of network reconnaissance means and the imposed constraints.

Haunted House: Physical Smart Home Event Verification in the Presence of Compromised Sensors

In this article, we verify physical events using data from an ensemble of smart home sensors. This approach both protects against event sensor faults and sophisticated attackers. To validate our system’s performance, we set up a “smart home” in an office environment. We recognize 22 event types using 48 sensors over the course of two weeks. Using data from the physical sensors, we verify the event stream supplied by the event sensors to detect both masking and spoofing attacks. We consider three threat models: a zero-effort attacker, an opportunistic attacker, and a sensor-compromise attacker who can arbitrarily modify live sensor data. For spoofed events, we achieve perfect classification for 9 out of 22 events and achieve a 0% false alarm rate at a detection rate exceeding 99.9% for 15 events. For 11 events the majority of masking attacks can be detected without causing any false alarms. We also show that even a strong opportunistic attacker is inherently limited to spoofing few select events and that doing so involves lengthy waiting periods. Finally, we demonstrate the vulnerability of a single-classifier system to compromised sensor data and introduce a more secure approach based on sensor fusion.

Dynamic Performance Analysis of STEP System in Internet of Vehicles Based on Queuing Theory.

The Internet of vehicles (IoV) is an important research area of the intelligent transportation systems using Internet of things theory. The complex event processing technology is a basic issue for processing the data stream in IoV. In recent years, many researchers process the temporal and spatial data flow by complex event processing technology. Spatial Temporal Event Processing (STEP) is a complex event query language focusing on the temporal and spatial data flow in Internet of vehicles. There are four processing models of the event stream processing system based on the complex event query language: finite automata model, matching tree model, directed acyclic graph model, and Petri net model. In addition, the worst-case response time of the event stream processing system is an important indicator of evaluating the performance of the system. Firstly, this paper proposed a core algorithm of the temporal and spatial event stream processing program based on STEP by Petri net model. Secondly, we proposed a novel method to estimate the worst-case response time of the event stream processing system, which is based on stochastic Petri net and queuing theory. Finally, through the simulation experiment based on queuing theory, this paper proves that the data stream processing system based on STEP has good dynamic performance in processing the spatiotemporal data stream in Internet of vehicles.

MICAR: multi-inhabitant context-aware activity recognition in home environments

The sensor-based recognition of Activities of Daily Living (ADLs) in smart-home environments enables several important applications, including the continuous monitoring of fragile subjects in their homes for healthcare systems. The majority of the approaches in the literature assume that only one resident is living in the home. Multi-inhabitant ADLs recognition is significantly more challenging, and only a limited effort has been devoted to address this setting by the research community. One of the major open problems is called data association, which is correctly associating each environmental sensor event (e.g., the opening of a fridge door) with the inhabitant that actually triggered it. Moreover, existing multi-inhabitant approaches rely on supervised learning, assuming a high availability of labeled data. However, collecting a comprehensive training set of ADLs (especially in multiple-residents settings) is prohibitive. In this work, we propose MICAR: a novel multi-inhabitant ADLs recognition approach that combines semi-supervised learning and knowledge-based reasoning. Data association is performed by semantic reasoning, combining high-level context information (e.g., residents’ postures and semantic locations) with triggered sensor events. The personalized stream of sensor events is processed by an incremental classifier, that is initialized with a limited amount of labeled ADLs. A novel cache-based active learning strategy is adopted to continuously improve the classifier. Our results on a dataset where up to 4 subjects perform ADLs at the same time show that MICAR reliably recognizes individual and joint activities while triggering a significantly low number of active learning queries.

VMV-GCN: Volumetric Multi-View Based Graph CNN for Event Stream Classification

Event cameras can perceive pixel-level brightness changes to output asynchronous event streams, and have notable advantages in high temporal resolution, high dynamic range and low power consumption for challenging vision tasks. To apply existing learning models on event data, many researchers integrate sparse events into dense frame-based representations which can work with convolutional neural networks directly. Although these works achieve high performance on event-based classification, their models need lots of parameters to process dense event frames which do not fit with the sparsity of event data. To utilize the sparse nature of events, we propose a voxel-wise graph learning model ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VMV-GCN</i> ) for spatio-temporal feature learning on event streams. Specifically, we design the volumetric multi-view fusion module ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VMVF</i> ) to extract spatial and temporal information from views of voxelized event data. Then we take representative event voxels as vertices and use a novel dual-graph construction strategy to connect them. By aggregating neighborhood information based on relationships of vertices, the proposed dynamic neighborhood feature learning module ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DNFL</i> ) can capture discriminative spatio-temporal features on dynamically updated graphs. Experiments show that our method achieves state-of-the-art performance with low model complexity on event-based classification tasks, such as object classification and action recognition.

Adaptive Slicing Method of the Spatiotemporal Event Stream Obtained from a Dynamic Vision Sensor.

The dynamic vision sensor (DVS) measures asynchronously change of brightness per pixel, then outputs an asynchronous and discrete stream of spatiotemporal event information that encodes the time, location, and sign of brightness changes. The dynamic vision sensor has outstanding properties compared to sensors of traditional cameras, with very high dynamic range, high temporal resolution, low power consumption, and does not suffer from motion blur. Hence, dynamic vision sensors have considerable potential for computer vision in scenarios that are challenging for traditional cameras. However, the spatiotemporal event stream has low visualization and is incompatible with existing image processing algorithms. In order to solve this problem, this paper proposes a new adaptive slicing method for the spatiotemporal event stream. The resulting slices of the spatiotemporal event stream contain complete object information, with no motion blur. The slices can be processed either with event-based algorithms or by constructing slices into virtual frames and processing them with traditional image processing algorithms. We tested our slicing method using public as well as our own data sets. The difference between the object information entropy of the slice and the ideal object information entropy is less than 1%.

Continuous discovery of Causal nets for non-stationary business processes using the Online Miner

Capturing business process specifics using a model is essential to effectively manage, control, and instruct the process participants with their roles and tasks. A normative process model is an invaluable source of information, not only for human inspection but also for software supporting and controlling the process. The actual process execution likely deviates from the normative model and the magnitude of deviation usually increases over time due to process evolution. A descriptive process model is far more informative when comes to the analysis of the actual execution of the process and spotting the deviations from the norm. However, handcrafting the descriptive model using process-related documents is prohibitively laborious. To automate the discovery of the descriptive process models, we propose Online Miner (OM), an algorithm that continuously builds process models using a stream of events raised by the process. OM first builds a sound process model using a historical event log and then incrementally adapts this model to new data chunks. OM represents the model using a Causal-net – a dependency graph of activities in which arcs are bound with each other to reflect parallel dependencies and exclusive alternatives. OM finds the maximal bindings being consistent with the event log. OM produces sound and perfect fit and generalizing well models. OM quickly reacts to the concept drift spot in the event log by seamlessly adapting the model. OM helps business analytics experts with their everyday tasks of monitoring, auditing, and enhancing business processes.

Mathematical Criteria for a Priori Performance Estimation of Activities of Daily Living Recognition.

Monitoring Activities of Daily Living (ADL) has become a major occupation to respond to the aging population and prevent frailty. To do this, the scientific community is using Machine Learning (ML) techniques to learn the lifestyle habits of people at home. The most-used formalism to represent the behaviour of the inhabitant is the Hidden Markov Model (HMM) or Probabilistic Finite Automata (PFA), where events streams are considered. A common decomposition to design ADL using a mathematical model is Activities–Actions–Events (AAE). In this paper, we propose mathematical criteria to evaluate a priori the performance of these instrumentations for the goals of ADL recognition. We also present a case study to illustrate the use of these criteria.

Online correlation for unlabeled process events: A flexible CEP-based approach

Process mining is a sub-field of data mining that focuses on analyzing timestamped and partially ordered data. This type of data is commonly called event logs. Each event is required to have at least three attributes: case ID, task ID/name, and timestamp to apply process mining techniques. Thus, any missing information need to be supplied first. Traditionally, events collected from different sources are manually correlated. While this might be acceptable in an offline setting, this is infeasible in an online setting. Recently, several use cases have emerged that call for applying process mining in an online setting. In such scenarios, a stream of high-speed and high-volume events continuously flow, e.g. IoT applications, with stringent latency requirements to have insights about the ongoing process. Thus, event correlation must be automated and occur as the data is being received.We introduce an approach that correlates unlabeled events received on a stream. Given a set of start activities, our approach correlates unlabeled events to a case identifier. Our approach is probabilistic. That implies a single uncorrelated event can be assigned to zero or more case identifiers with different probabilities. Moreover, our approach is flexible. That is, the user can supply domain knowledge in the form of constraints that reduce the correlation space. This knowledge can be supplied while the application is running. We realize our approach using complex event processing (CEP) technologies. We implemented a prototype on top of Esper, a state of the art industrial CEP engine. We compare our approach to baseline approaches. The experimental evaluation shows that our approach outperforms the throughput and latency of the baseline approaches. It also shows that using real-life logs, the accuracy of our approach can compete with the baseline approaches.

Seasonal flashiness and high frequency discharge events in headwater streams in the North Carolina Piedmont (United States)

AbstractFirst‐order streams are rarely the focus of research even though they play an important hydrological and ecological role in upland landscapes and may be more susceptible to human impacts than larger streams. We studied 16 first‐order streams in the North Carolina Piedmont to examine relationships between land use, discharge, flashiness and conductivity in the headwaters of the Rocky River and McDowell Creek. Stream gauges were installed on each of the 16 headwater streams to provide constant monitoring of stream depth, temperature and specific conductance. Rating curves were created for each site using models that combined field measured and Manning's calculated discharge. Initial attempts to measure flashiness using the Richards‐Baker Index (RBI) indicated that it is not a good proxy for flashiness in small streams because its use of median daily flows do not capture fluctuations that occur over short time periods. Instead, we developed a new method that focused on the number of days each stream flows above its 99th percentile flow (DMQ99). We also examined individual storm events across multiple seasons. We found that headwater streams in the Piedmont are flashier than more commonly gauged second and third order streams because first‐order streams are more directly connected to the landscape. Streams located in suburban areas tend to be flashier than streams in forested areas as a result of the increase in impervious surfaces and the reduction of evapotranspiration (ET) caused by development. Additionally, we found that all streams become flashier in the winter when ET is lower, suggesting important feedbacks between physical and biological systems. All streams experience periodic extreme conductivity events, indicating significantly impaired ecological conditions. More developed basins generally have higher maximum specific conductance compared with less developed basins although streams have significant natural variability as well. Overall, our results indicate that first‐order streams in the Piedmont are highly impacted by development and that extra caution should be exercised when altering landscapes near small streams.

New Findings Relating Tidal Variability and Solar Activity in the Low Latitude MLT Region

AbstractMesospheric winds have been measured by meteor radar at Cachoeira Paulista (22.7°S; 45°W) since April 1999. The tidal components were analyzed over about 21 years of available data exhibiting an annual and semi‐annual variation. Amplitudes of meridional diurnal (semidiurnal) tide are on averaged 30% (28%) and the zonal ones are 14% (20%) stronger at solar minimum than at solar maximum. The anticorrelation between F10.7 cm solar flux and amplitudes of the semiannual oscillation of diurnal and semidiurnal tides is presented. Additionally, the sporadic E (Es) layers occurrence has an anti‐correlation with solar activity due to the tidal wind variation. A discussion about the physical mechanism is performed in terms of the particle precipitation during High‐Speed Stream (HSS) events according to the solar cycle. Finally, a superposed epoch analysis of the tidal amplitudes during the HSS events in 2018 is presented. And a slight increase in all tidal components is seen when the structure reached the Earth and in the following days showing that indeed the electron precipitation during HSS events affect the tidal amplitudes.

Tuning Resistive Switching Behavior by Controlling Internal Ionic Dynamics for Biorealistic Implementation of Synaptic Plasticity

AbstractMemristive devices have demonstrated rich switching behaviors that closely resemble synaptic functions and provide a building block to construct efficient neuromorphic systems. It is demonstrated that resistive switching effects are controlled not only by the external field, but also by the dynamics of various internal state variables that facilitate the ionic processes. The internal temperature, for example, works as a second‐state variable to regulate the ion motion and provides the internal timing mechanism for the native implementation of timing‐ and rate‐based learning rules such as spike timing dependent plasticity (STDP). In this work, it is shown that the 2nd state‐variable in a Ta2O5‐based memristor, its internal temperature, can be systematically engineered by adjusting the material properties and device structure, leading to tunable STDP characteristics with different time constants. When combined with an artificial post‐synaptic neuron, the 2nd‐order memristor synapses can spontaneously capture the temporal correlation in the input streaming events.

Adaptive Event Address Map Denoising for Event Cameras

Event cameras provide high temporal resolution and high dynamic range and have application prospects in areas such as visual navigation, robotics, and image reconstruction. Event cameras output an asynchronous stream of address events that can be highly polluted by noise due to lighting and motion conditions. In order to better analyze and process noise with different spatiotemporal correlation, we introduce a method using an event address map (EAM) representation and classify event noise into background noise, near-edge noise, and recessive noise. In addition, since near-edge noise and recessive noise have a closer spatiotemporal correlation with real events, they are difficult to suppress with current denoising methods. Accordingly, we propose an adaptive EAM denoising method by adjusting the time window to form the EAM by measuring the number of events after pre-denoising and performing median filtering and opening. The method can better adapt to scenes of different motions to process near-edge noise, and can generate events to fill recessive noise. The output EAM can be directly used for subsequent process. Alternatively, to provide high temporal resolution, we restore the stream of address events from the output EAM. To suitably evaluate event denoising performance on real-world data, we introduce the average local event variance and amount of recessive noise, and the proposed method achieves improvements of 24.94% and 72.79%, respectively, compared with conventional denoising.

Assessing accessibility: an instrumental case study of a community music group

ABSTRACT Creating accessible events is a pressing issue for many music organisations. In the United States, the term accessibility has strong ties to disability, and it is an important concept because what is deemed accessible directly impacts who is included and excluded from music experiences. Music Community Lab (MCL) runs a series of events in New York City called Monthly Music Hackathon NYC. They aim to promote ‘diversity across backgrounds, perspectives, and abilities’. This instrumental case study sought to examine how MCL participants conceptualise accessibility as well as analyze participants’ suggestions for improving the accessibility of MCL events. Sixty-two people who attended one of three MCL events completed a demographic survey and 57 of those respondents participated in an interview. Findings reveal that 63% (n = 36) of participants associated accessibility with inclusivity and 35% (n = 20) of participants associated accessibility with ease of access to resources, places, and experiences. Participants’ suggestions for improving accessibility included social media marketing (n = 23; 40%) and ease of access approaches (n = 11; 19%) including CART, ASL, and live streaming events. Accessibility is challenging for community music groups like MCL to navigate because it is a complex construct with varied interpretations.

Threat Analysis and Distributed Denial of Service (DDoS) Attack Recognition in the Internet of Things (IoT)

The Internet of Things (IoT) plays a crucial role in various sectors such as automobiles and the logistic tracking medical field because it consists of distributed nodes, servers, and software for effective communication. Although this IoT paradigm has suffered from intrusion threats and attacks that cause security and privacy issues, existing intrusion detection techniques fail to maintain reliability against the attacks. Therefore, the IoT intrusion threat has been analyzed using the sparse convolute network to contest the threats and attacks. The web is trained using sets of intrusion data, characteristics, and suspicious activities, which helps identify and track the attacks, mainly, Distributed Denial of Service (DDoS) attacks. Along with this, the network is optimized using evolutionary techniques that identify and detect the regular, error, and intrusion attempts under different conditions. The sparse network forms the complex hypotheses evaluated using neurons, and the obtained event stream outputs are propagated to further hidden layer processes. This process minimizes the intrusion involvement in IoT data transmission. Effective utilization of training patterns in the network successfully classifies the standard and threat patterns. Then, the effectiveness of the system is evaluated using experimental results and discussion. Network intrusion detection systems are superior to other types of traditional network defense in providing network security. The research applied an IGA-BP network to combat the growing challenge of Internet security in the big data era, using an autoencoder network model and an improved genetic algorithm to detect intrusions. MATLAB built it, which ensures a 98.98% detection rate and 99.29% accuracy with minimal processing complexity, and the performance ratio is 90.26%. A meta-heuristic optimizer was used in the future to increase the system’s ability to forecast attacks.

Differentially private frequent episode mining over event streams

Frequent episode mining is a wide range framework of data mining from sequential data with many applications, which is a totally short-ordered collection of event-types and unearths temporal correlations without information loss over event streams. While offering substantial benefits, directly releasing frequent episodes to the public will enormously threaten the individual’s privacy. However, there is little work so far concentrating on privately frequent episode mining. In this paper, we investigate the privacy problem in mining frequent episodes from event streams due to continuous releases in successive windows and propose a real-time differentially private frequent episode mining algorithm over event streams to avoid the privacy leakage with ω-event privacy guarantee. To obtain private frequent episodes, we propose a sample-based perturbation approach, which improves the accuracy of selecting frequent episodes based on sampling databases. To reduce the privately mining time and avoid repeatedly privacy budget allocation to coincident window of adjacent releases as much as possible, we present an incremental perturbation approach according to the judgment in dissimilarity calculation mechanism. Meanwhile, in order to protect data collected from any ω successive timestamps over event streams, we employ an adaptive ω-event privacy mechanism on the basis of the dynamicity of episodes. Finally, experimental results on real-world datasets demonstrate the effectiveness and efficiency of our algorithm.

An analytical, fully relativistic framework for tidal disruption event streams in Schwarzschild geometry

ABSTRACT We present an analytical and fully relativistic framework for studying the self-intersection of tidal disruption event (TDE) streams, restricting ourselves to the Schwarzschild spacetime. By taking advantage of the closed-form solution to the geodesic equations in the Schwarzschild metric, we calculate properties of the self-intersection without numerically evaluating the geodesic equations or making any post-Newtonian approximations. Our analytical treatment also facilitates geometric definitions of the orbital semimajor axis and eccentricity, as opposed to Newtonian formulas which lead to unphysical results for highly relativistic orbits. Combined with assumptions about energy dissipation during the self-intersection shock, our framework enables the calculation of quantities such as the fraction of material unbound during the self-intersection shock, and the characteristic semimajor axes and eccentricities of the material that remains in orbit after the collision. As an example, we calculate grids of post-intersection properties in stellar and supermassive black hole (SMBH) masses for disruptions of main-sequence stars, identifying regions where no material is ejected during self-intersection (e.g. SMBH mass $\lesssim 5\times 10^6\, {\rm M_\odot }$ for $1\, {\rm M_\odot }$ stars disrupted at the tidal radius), potentially explaining the TDEs observed by SGR/eROSITA that are visible in X-rays but not optical wavelengths. We also identify parameters for which the post-intersection accretion flow has low eccentricity (e ≲ 0.6), and find that the luminosity generated by self-intersection shocks only agrees with observed trends in the relationship between light curve decay time-scales and peak luminosities over a narrow range of SMBH masses.

NeuroIV: Neuromorphic Vision Meets Intelligent Vehicle Towards Safe Driving With a New Database and Baseline Evaluations

Neuromorphic vision sensors such as the Dynamic and Active-pixel Vision Sensor (DAVIS) using silicon retina are inspired by biological vision, they generate streams of asynchronous events to indicate local log-intensity brightness changes. Their properties of high temporal resolution, low-bandwidth, lightweight computation, and low-latency make them a good fit for many applications of motion perception in the intelligent vehicle. However, as a younger and smaller research field compared to classical computer vision, neuromorphic vision is rarely connected with the intelligent vehicle. For this purpose, we present three novel datasets recorded with DAVIS sensors and depth sensor for the distracted driving research and focus on driver drowsiness detection, driver gaze-zone recognition, and driver hand-gesture recognition. To facilitate the comparison with classical computer vision, we record the RGB, depth and infrared data with a depth sensor simultaneously. The total volume of this dataset has 27360 samples. To unlock the potential of neuromorphic vision on the intelligent vehicle, we utilize three popular event-encoding methods to convert asynchronous event slices to event-frames and adapt state-of-the-art convolutional architectures to extensively evaluate their performances on this dataset. Together with qualitative and quantitative results, this work provides a new database and baseline evaluations named NeuroIV in cross-cutting areas of neuromorphic vision and intelligent vehicle.

The Impact of Forest Fertilization on the Ecological Quality of Two Hemiboreal Streams

The present study aimed to detect any changes in concentrations of nutrients and evaluate the impact on the quality of two hemiboreal streams that collect a discharge from two fertilized Scots pine stands. In 2017, nitrogen-containing mineral fertilizer was spread in pine stands on mineral soil located near the first stream. In 2018, potassium containing wood ash was spread in pine stands on organic soil near the second stream. From 2017 to 2020, surveys of physico-chemical parameters, diatoms, macrophytes, and macroinvertebrates were performed to determine the possible effects of fertilization on the ecological quality of the streams. A control site upstream of the fertilized forest stand and a treatment site downstream of the fertilized forest stand was monitored at each stream. Water quality indices, chemical parameters of surface water, and indicator species analysis showed no short-term impact of forest soil improvement with wood ash and ammonium nitrate. We found no clear patterns before and after the fertilization events in both streams, although we did observe inter- and intra-annual differences in aquatic biota and stream ecological quality mainly caused by local environmental factors.