Find research datasets worth reusing

opencc-by-4.0Dec 2021View details →

Bus Violence: a large-scale benchmark for video violence detection in public transport

Dataset The Bus Violence dataset is a large-scale collection of videos depicting violent and non-violent situations in public transport environments. This benchmark was gathered from multiple cameras located inside a moving bus where several people simulated violent actions, such as stealing an object from another person, fighting between passengers, etc. It contains 1,400 video clips manually annotated as having or not violent scenes, making it one of the biggest benchmarks for video violence detection in the literature. Specifically, videos are recorded from three cameras at 25 Frames Per Second (FPS) --- two cameras located in the corners of the bus (with resolution 960x540 px) and one fisheye in the middle (1280x960 px). The clips have a minimum length of 16 frames and a maximum of 48 frames, capturing a very precise action (either violence or non-violence). The dataset is perfectly balanced, containing 700 videos of violence and 700 videos of non-violence. The Bus Violence dataset is intended as a test data benchmark. However, for researchers interested in using our data also for training purposes, we provide training and test splits. In this repository, we provide <ul> <li> the 1,400 video clips divided into two folders named Violence /NoViolence, containing clips of violent situations and non-violent situations, respectively; </li> <li> two txt files containing the names of the videos belonging to the training and test splits, respectively. </li> </ul>   Citing our work If you found this dataset useful, please cite the following paper <blockquote> <pre>@inproceedings{bus_violence_dataset_2022,   title = {Bus Violence: An Open Benchmark for Video Violence Detection on Public Transport}, doi = {10.3390/s22218345},   url = {https://doi.org/10.3390%2Fs22218345},   year = 2022,   month = {oct},   publisher = {{MDPI} {AG}},   volume = {22},   number = {21},   pages = {8345},   author = {Luca Ciampi and Pawe{\l} Foszner and Nicola Messina and Micha{\l} Staniszewski and Claudio Gennaro and Fabrizio Falchi and Gianluca Serao and Micha{\l} Cogiel and Dominik Golba and Agnieszka Szcz{\k{e}}sna and Giuseppe Amato}, journal = {Sensors} } </pre> </blockquote> and this Zenodo Dataset <blockquote> <pre>@dataset{pawel_bus_violence_zenodo, author = {Paweł Foszner, Michał Staniszewski, Agnieszka Szczęsna, Michał Cogiel, Dominik Golba, Luca Ciampi, Nicola Messina, Claudio Gennaro, Fabrizio Falchi, Giuseppe Amato, Gianluca Serao}, title = {{Bus Violence: a large-scale benchmark for video violence detection in public transport}},   month = sep,   year = 2022,   publisher = {Zenodo},   version = {1.0.0},   doi = {10.5281/zenodo.7044203}, url = {https://doi.org/10.5281/zenodo.7044203} } </pre> </blockquote>   Contact Information Blees Sp. z o.o., Gliwice, Poland mstaniszewski@blees.co   Acknowledgments The presented dataset was supported by: European Union funds awarded to Blees Sp. z o.o. under grant POIR.01.01.01-00-0952/20-00 “Development of a system for analysing vision data captured by public transport vehicles interior monitoring, aimed at detecting undesirable situations/behaviours and passenger counting (including their classification by age group) and the objects they carry”); EC H2020 project "AI4media: a Centre of Excellence delivering next generation AI Research and Training at the service of Media, Society and Democracy" under GA 951911; research project INAROS (INtelligenza ARtificiale per il mOnitoraggio e Supporto agli anziani), Tuscany POR FSE CUP B53D21008060008.   License The Bus Violence dataset was acquired by Blees Sp. z o.o. and is released under a Creative Commons Attribution license for non-commercial use.

opencc-by-4.0May 2019View details →

Vehicle CAN bus data

The dataset contains 20Hz sampled CAN bus data from a passenger vehicle, e.g. WheelSpeed FL (speed of the front left wheel), SteerAngle (steering wheel angle), Role, Pitch, and accelerometer values per direction. Due to an export error, all GPS data is currently 0, but we are currently looking for a solution and will update the record as soon as possible.

opencc-by-4.0Sep 2021View details →

Synthetic bus validation of Rennes public transportation network focus on Beaulieu campus

Synthetic data based on the STAR/Keolis open data version GTFS_2021.1.0.3_20210830_20211017 for regular weeks and GTFS_2020.10.1_20210705_20210829 for holidays. This dataset contains the generation of bus validations for 3500 individuals during 14 weeks with an average of 15 travels per week. It also contains GPS information of the bus stops associated to the validation.

opencc-by-4.0Nov 2022View details →

IEEE New England 39-bus test case: Dataset for the Transient Stability Assessment

The dataset contains 350 features engineered from the phasor measurements (PMU-type) signals from the IEEE New England 39-bus power system test case network, which are generated from the 9360 systematic MATLAB®/Simulink electro-mechanical transients simulations. It was prepared to serve as a convenient and open database for experimenting with different types of machine learning techniques for transient stability assessment (TSA) of electrical power systems. Different load and generation levels of the New England 39-bus benchmark power system were systematically covered, as well as all three major types of short-circuit events (three-phase, two-phase and single-phase faults) in all parts of the network. The consumed power of the network was set to 80%, 90%, 100%, 110% and 120% of the basic system load levels. The short-circuits were located on the busbar or on the transmission line (TL). When they were located on a TL, it was assumed that they can occur at 20%, 40%, 60%, and 80% of the line length. Features were obtained directly from the time-domain signals at the pickup time (pre-fault value) and at the trip time (post-fault value) of the associated distance protection relays. This is a stochastic dataset of 3120 cases, created from the population of 9360 systematic simulations, which features a statistical distribution of different fault types, as follows: single-phase (70%), double-phase (20%) and three-phase faults (10%). It also features a class imbalance, with less than 20% of cases belonging to the unstable class. Dataset is a compressed CSV file. List of feature names in the dataset: <ul> <li>WmGx - rotor speed for each generator Gx, from G1 to G10,</li> <li>DThetaGx - rotor angle deviation for each generator Gx, from G1 to G10,</li> <li>ThetaGx - rotor mechanical angle for each generator Gx, from G1 to G10,</li> <li>VtGx - stator voltage for each generator Gx, from G1 to G10,</li> <li>IdGx - stator d-component current for each generator Gx, from G1 to G10,</li> <li>IqGx - stator q-component current for each generator Gx, from G1 to G10,</li> <li>LAfvGx - pre-fault power load angle for each generator Gx, from G1 to G10,</li> <li>LAlvGx - post-fault power load angle for each generator Gx, from G1 to G10,</li> <li>PfvGx - pre-falut value of the generator active power for each generator Gx, from G1 to G10,</li> <li>PlvGx - post-falut value of the generator active power for each generator Gx, from G1 to G10,</li> <li>QfvGx - pre-falut value of the generator reactive power for each generator Gx, from G1 to G10,</li> <li>QlvGx - post-falut value of the generator reactive power for each generator Gx, from G1 to G10,</li> <li>VAfvBx - pre-fault bus voltage magnitude in phase A for each bus Bx, from B1 to B39,</li> <li>VBfvBx - pre-fault bus voltage magnitude in phase B for each bus Bx, from B1 to B39,</li> <li>VCfvBx - pre-fault bus voltage magnitude in phase C for each bus Bx, from B1 to B39,</li> <li>VAlvBx - post-fault bus voltage magnitude in phase A for each bus Bx, from B1 to B39,</li> <li>VBlvBx - post-fault bus voltage magnitude in phase B for each bus Bx, from B1 to B39,</li> <li>VClvBx - post-fault bus voltage magnitude in phase C for each bus Bx, from B1 to B39,</li> <li>Stability - binary indicator (0/1) that determines if the power system was stable or unstable (0 - stable, 1 - unstable); this is the label variable.</li> </ul> License: Creative Commons CC-BY. Disclaimer: This dataset is provided "as is", without any warranties of any kind.

opencc-by-4.0Nov 2013View details →

Lappeenranta City Bus Cycle

Report of Lappeenranta city bus cycle: Lappeenranta Route 1 Address of Description document: http://www.doria.fi/bitstream/handle/10024/93685/NBNfi-fe201311117322.pdf?sequence=3

The P2P-IEEE 14 bus system data set

This data set models the IEEE 14-bus system for studies on P2P electricity markets, including real data of consumption, solar and wind power from Australia. This data set is characterized by 30 minutes time-step over one year, i.e. from July 2012 to June 2013. The transmission system comprises 14 buses and 20 lines, and its characteristics are based on [1]. The original number of generators was increased to 8 generators, i.e. 1 coal-based generator, 2 gas-based generators, 3 wind turbines and 2 PV plants. The data set uses the original number of 11 loads. The bus 1 represents the upstream connection to the main grid, where the generator assumes an infinite power. The market price from the Australian Energy Market Operator is used in this generator. It is assumed the same period from July 2012 to June 2013 [4]. This data set supposes a tariff of 10$/MWh for using the main grid. The energy imported and exported in bus 1 has to account this extra cost. Thus, the exportation price is equal to the market price minus this grid tariff. On the other hand, the importation price is equal to the market price plus this grid tariff. The wind production has been based on the data set from [2]. The time resolution has been converted from 5 minutes to 30 minutes. The authors would like to acknowledge that the data set in [2] was processed by Stefanos Delikaraoglou and Jethro Dowell. The solar production and load consumption are taken from [3]. The load consumption is split into fixed and flexible consumption per time-step. Since there is no access to the total capacity of the flexible consumption, we split the daily flexible consumption over each time-step. In this way, the maximum consumption is equal to the fixed consumption plus twice this flexible consumption per time-step. The minimum consumption is equal to the fixed consumption in each time-step. The wind, solar and load data sets have been normalized, i.e. values relative to rated power. Then, these normalized sequences were multiplied by the capacity of each element. The data is intended for use in studies related to consumer-centric electricity markets, e.g.: <ul> <li>Validate new market designs or business models;</li> <li>Assess the impact of new grid operation strategies;</li> <li>Test the effect of strategic behavior by producers or consumers.</li> </ul>

Using the Genetic Algorithm for the Optimization of Dynamic School Bus Routing Problem-Figure 9. Step by step drawing of the selected route solution

When the solution that is desired to be viewed is double clicked, the connections between bus stops are drawn in turn, and the route is shown as can be seen in Figure 9.

Using the Genetic Algorithm for the Optimization of Dynamic School Bus Routing Problem-Figure 7. Application main form which includes route, bus stops and GA parameters

For this study, the school bus routes within the Ankara Province were used as case studies. The school bus routes were recorded by using the Android application and instantaneous GPS monitoring method. The home address of each student was taken as a stopping point. At the end of each route, the distances between the beginning and end points were recorded. After transferring the obtained route data into the database, the ill-adapted points were eliminated. By means of the developed application, the existing school bus routes are dynamically optimized using GA. It was developed both as a mobile and desktop application.  Using Android- based mobile software, the information regarding GPS locations of bus stops and school buses is transferred to the server on a real-time basis. Using the desktop software, where GAs are run, these coordinates are shown on a Google map, and the most suitable route is produced and sent to the school bus via server. This method provides the opportunity to dynamically reflect certain factors such as a different initial point for the school bus, some students being absent from the school on a particular day, and eventual changes on the existing roads on the route production process.

Using the Genetic Algorithm for the Optimization of Dynamic School Bus Routing Problem-Figure 6. Mutation process

The individuals obtained at the end of crossing over might not provide the desired level of variability. In that case, the produced individuals are mutated independently from another individual in such a way that their own gene sequence will change. The mutation process is performed in the event that the mutation possibility that is specified in the beginning comes true. The results obtained from mutation can enhance the outcome or make it worse. It is of utmost importance to specify the most suitable mutation possibility. This possibility should be high enough to prevent the method from becoming stuck at a local point, but at the same time, low enough to allow the best results produced by crossing over and multiplexing. In this study, the mutation possibility was selected as 10%, and the locations of two randomly selected bus stops were changed during the mutation process. As in the crossing over, also during this process, the limitations regarding producing a new individual (route) were adapted. Figure 6 shows an example to mutation process.

Using the Genetic Algorithm for the Optimization of Dynamic School Bus Routing Problem-Figure 4. Example of a chromosome structure with permutation coding

Each chromosome found in the population formed in the GA is structurally an equal-length coded series. The chromosomes are made of genes. For coding purposes, binary, permutation, and value coding methods are widely used. In the travelling salesman or other similar VRPs, permutation coding technique is preferred over the other techniques. Using the permutation coding technique, each chromosome found in the population is expressed in terms of the numbers of each stop to be followed in the route, as shown in Figure 4.

Using the Genetic Algorithm for the Optimization of Dynamic School Bus Routing Problem-Figure 3. The flow diagram of the SBRP solution by using GA

The distance optimization needed for the formation of the objective function that can be seen in equation number 1 was done using GA operators and parameters. The flowchart that can be seen in Figure 3 shows how the school bus routes are formed using GA.

Using the Genetic Algorithm for the Optimization of Dynamic School Bus Routing Problem-Figure 10. Obtained experimental results

The GA procedure was carried out using the parameter values that can be seen in Table 1, for a total of 10 school bus services’ routes identified by means of mobile-based software for a school located in the Ankara Province. The obtained experimental results are shown in Table 2 and Figure 10. The algorithm working duration also includes the formation of a distances’ matrix.

Using the Genetic Algorithm for the Optimization of Dynamic School Bus Routing Problem-Figure 1. Example of a VRP solution

VRP is widely described as the process where the distribution vehicles deliver goods in a depot to the clients found in geographically dispersed locations and then return to the depot on the optimum route. The VRP, first introduced by Dantzig and Ramser in 1959, aims to minimize the total distance to be covered during the routing of the vehicles in a centrally-located depot (Dantzig & Ramser, 1959). The routing process is operated by taking the vehicle capacities into consideration so as to ensure that each client located on the routing plan is visited only once. The depot, vehicle, and goods concepts mentioned in the general description of VRP can be re-adapted as school, student and bus stop, based on different sectors that the problem applies to. Figure 1 shows a scheme indicating the VRP solution applied to a distribution system, where the clientele network is dispersed over three different zones.

Using the Genetic Algorithm for the Optimization of Dynamic School Bus Routing Problem-Figure 8. Listing of obtained solutions after running GA on the route

During the route planning, school is the initial point if the students are going home from school, whereas it is the final point if they are going to the school from their homes. The application provides the opportunity to fix not only the school, but also the bus stops as initial or final points. Since the authors hope to develop a routing solution for more than one school in a future study, both the initial and final bus stops were given the opportunity to be selected so that the school bus can complete distribution/collection duties for one school and can go to another school for routing. When these selections are made, the initial and final genes of the chromosomes produced in the population were fixed through these selections. When the TSP option is on, on the other hand, the school bus returning to the initial point after completing the distribution/collection duties is included in the routing process. The information regarding solutions, route distances and the number of iterations produced after all the parameters are defined and the relevant selections are made are listed as shown in Figure 8. Under the solutions list, total crossover, total number of mutations, and algorithm working times are shown.

Using the Genetic Algorithm for the Optimization of Dynamic School Bus Routing Problem-Figure 5. Crossover process

In the selection mechanism, the individuals passed down from the previous generation occasionally cannot produce a better individual. In that case, the compatibility of the individuals might worsen, while producing the exact opposite is what is expected. To avoid this, the elitism operator is used and it is ensured that the best individual of the previous generation is passed down to the next generation, even though the current population is diminishing on an overall basis as a result of the production operators (Goldberg, 1989). In the current study, the consecutive selection method and elitism selection were preferred. For this aim, after calculating the compatibility function, the population was ranked according to the population function values (total route length). In case the crossover possibility is realized, the number of individuals to select will be determined according to the parameter related to the crossover size. To ensure a high level of variability in the generation, it is suggested that this possibility is taken as 50% and 95% (Goldberg, 1989). The crossover process allows the production of a new individual using the genes taken from two individuals, based on the selected crossover method. In this study, a single point crossover method was selected. During the crossing over, limitations that were previously mentioned regarding the formation of a new initial population were taken into consideration. The identified initial or final point was fixed and kept out of the context of crossing over. An example of crossover can be seen in Figure 5.

Using the Genetic Algorithm for the Optimization of Dynamic School Bus Routing Problem-Figure 2. A dynamic school bus routing case

VRP’s can also be classified into two categories as dynamic and static routing problems. In the static VRP’s, the stops/locations that the vehicle will visit are pre-specified and do not change during the distribution/collection process. In dynamic VRP’s, on the other hand, new stops can be added to the planned route during the process or certain stops can be omitted. In similar dynamic problems, some or all of the access points are not known in the beginning. These points are dynamically defined during the route design or planning stages. In the dynamic VRP, using a real- time communication network between the vehicle and decision-making system, the vehicle routes can be re-defined during the operation. This type of problems is defined as online or real-time problems by some scholars (Pillac, Gendreau, Guéret & Medaglia, 2013). Two examples of this can be certain orders getting cancelled or new orders being taken while a water distribution vehicle is on its route, or a school bus being informed on its route that certain students will be absent from school that day.  In current conditions, dynamic VRP’s are more frequently needed, and are attributed with a more specific importance. The first study dealing with dynamic VRP was carried out by Wilson and Colvin (Pillac, Gendreau, Guéret & Medaglia, 2013). The enhancements in GPS, traffic sensors, and mobile communication systems caused a further acceleration in studies carried out in this field. Within the context of this study, DSBRP will be investigated. DSBRP is graphically explained in Figure 2.

opencc-by-4.0Jul 2019View details →

Automotive CAN bus data: An Example Dataset from the AEGIS Big Data Project

Here you find an example research data dataset for the automotive demonstrator within the "AEGIS -  Advanced Big Data Value Chain for Public Safety and Personal Security" big data project, which has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 732189. The time series data has been collected during trips conducted by three drivers driving the same vehicle in Austria. The dataset contains 20Hz sampled CAN bus data from a passenger vehicle, e.g. WheelSpeed FL (speed of the front left wheel), SteerAngle (steering wheel angle), Role, Pitch, and accelerometer values per direction. GPS data from the vehicle (see signals 'Latitude_Vehicle' and 'Longitude_Vehicle' in h5 group 'Math') and GPS data from the IMU device (see signals 'Latitude_IMU', 'Longitude_IMU' and 'Time_IMU' in h5 group 'Math') are included. However, as it had to be exported with single-precision, we lost some precision for those GPS values.   For data analysis we use R and R Studio (https://www.rstudio.com/) and the library h5. e.g. check file with R code: library(h5) f <- h5file("file path/20181113_Driver1_Trip1.hdf") summary(f["CAN/Yawrate1"][,]) summary(f["Math/Latitude_IMU"][,]) h5close(f)