Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
6
datasets available to search
ShareScore release 0.9.0
Dataset results
6 results for “computer vision algorithms”
Super-resolving ocean dynamics from space with computer vision algorithms: training datasets
<p>We provide here the datasets used for the development of the dilated Adaptive Residual Network for the super-resolution of ocean Absolute Dynamic Topography described in <em>Buongiorno Nardelli et al.</em> (2022). The model is designed to combine satellite altimetry and thermal observations and provides super-resolved dynamic topography. The training/test datasets have been built starting from the data originally prepared for an Observing System Simulation Experiment carried out in the framework of the European Space Agency CIRCOL project [<em>Ciani et al.</em>, 2021]. They consist of one year of synthetic daily Absolute Dynamic Topography (ADT), surface geostrophic currents and sea surface temperature data obtained from Copernicus Marine Service Mediterranean Forecasting System (MFS) (Product ID: MEDSEA-ANALYSIS- FORECAST-PHY-006-013) [<em>Clementi et al. 2021</em>]. Synthetic Altimeter-derived ADT maps were obtained by first sampling the model output along the actual tracks of a synthetic constellation composed of 4 Radar Altimeters: Jason-3, Sentinel-3A, SARAL/Altika, and Cryosat-2 missions (this step is achieved by running the SWOT simulator software [<em>Gaultier et al.</em>, 2016]) and successively applying the DUACS (<em>Data Unification and Altimeter Combination System)</em> mapping method. The original input images cover the entire Mediterranean domain at 1/24° spatial resolution, leading to an individual image size of 380x1000 pixels. Here, we have randomly chosen 40 dates (~11% of the total) to be kept aside as fully independent test data, and successively re-sampled the original images extracting much smaller tiles (76x100), which are used as input to the network training. The tiles are extracted by going through a double loop on latitude and longitude, imposing a spatial overlap of 50%. Full details on data pre-processing (e.g.normalization strategies) are given in the paper:</p> <ul> <li>Buongiorno Nardelli, B.; Cavaliere, D.; Charles, E.; Ciani, D. Super-Resolving Ocean Dynamics from Space with Computer Vision Algorithms. <em>Remote Sens.</em>, <strong>2022</strong>, 14, 1159. https://doi.org/10.3390/rs14051159</li> </ul>
Tassie BRUV: A benchmark data set for computer vision and movement quantification algorithms
Open the record for dataset details and reuse information.
Computer vision: algorithms to make sense of the world
<p><strong>The following video describes how computer vision is used by the ROMI platform for object and species detection in both 2D and 3D, and how it is integral to the weeding tool. Funded by EU Grant 773875.</strong></p> <p><em>Videos are available in:</em></p> <ul> <li>Hi-res (1080p Apple ProRes)</li> <li>Mid-res (1080p H265)</li> </ul> <p><strong>Video script:</strong></p> <p>(CAMPRODON) What's computer vision? Mmm … computer vision for me is making sense of pixels. I think computer vision has a profound effect in the way that we understand the world because as humans vision is so centric right. If dogs would be making computers, maybe they wouldn't talk that much about vision. But for us it's so centric in the way that we perceive the world and the way we learn about the world, that actually I think it's easier for that of course to program and think useful ways machines could get information you know through vision. And also especially because vision is one of the much more complex senses that we have.<br> <br> (SOLLAZZO) Image and videos that represent nowadays the 80 percent of the data that we produce and the introduction of computer vision and machine learning becomes necessary to start extrapolating information out of this new source of data.<br> <br> (COLLIAUX) So just a point of clarification because we often talk about AI and so just to be a bit more precise about what we do in ROMI. Because AI is quite a vague term, and so what we do mainly is robotics and computer vision.<br> <br> (SOLLAZZO) Computer vision is at the end a limited set of tools and systems that are basically based on mathematical representation and description of the pixel that represent the image, they are part of the image, and machine learning is based on a different approach of interpretation of those pixels.<br> <br> (COLLIAUX) So the rover is for weeding and to remove the weeds you need to detect the weeds first and so we use a computer vision algorithm to detect where are the weeds where are the salads.<br> <br> (SOLLAZZO) So let's see one by one which are the methods that we implemented, in our algorithm, in our system. So we start with the feature extraction in order to do that in fact we go one by one over the images and we understand which are the pixels in common between one and the other. From these method in fact it's possible to recreate an orthomosaic view, an orthomosaic image, but afterwards we need to align it to all the previous images that we've been creating in the previous analysis. So after the generation of the orthomosaic view, what we do is that we start to cut the main image into a portion into a series of smaller portions. This facilitates the execution of the machine learning algorithm and the possibility to recognise the presence or not, of lettuce in the scene. After the recognition has been performed we put together the images once again and we can reconstruct an orthomosaic view with a detected position of the different lettuce. This is necessary to understand not only the position but also the area of growth that these different lettuce are occupying over time. From the geolocation of every single plant we start to analyse the growing curve over time. This is possible thanks to the implementation of ‘Mask RCNN’. So thanks to the generation of all these different areas that during time, will tell us the growing pattern of every single lettuce, and this will be extremely useful to understand when is the moment to harvest the plant when the plant is in fact bolting, more or less this is ok.<br> <br> (COLLIAUX) So we do what I showed was about 2d computer vision, but we do a lot of 3d computer vision also in the project and so let me show you a bit what we do with a plant scanner. So it is uh used by biologists to study the geometry of the plants so they want to reconstruct the pre-architecture of a plant and study that architecture. So for this we take many images of a plant by turning a camera in a circle around the plant, we generate a mask but again a segmentation algorithm to detect where where the plant is and where the background is, and then we can generate a point cloud by an algorithm called ‘space carving’ or ‘shape from silhouette’ which based on the many silhouettes you collected it looks for it it carves the space for the shape which is the most compatible with all the projection of the shape.<br> <br> (CAMPRODON) So what we're doing in ROMI at the end, I would say in a way we hack existing technologies, we take advantage of the low cost cameras that exist in phones right we don't need to rely anymore in high-end industrial cameras, we take advantage of the low-cost computational power, computing cheaper than ever. So these images that we take we can process them with software in ways that was not possible before, we take advantage of software, of especially of open source software and free software and then we build the training models right, so this software is capable to detect on top of that images insights, to go from data to information.</p>
Computer Vision-Based Algorithm for Precise Defect Detection and Classification in Photovoltaic Modules
Open the record for dataset details and reuse information.
Filming the sound: Anomaly Detection on Audio Tape Recordings using Computer Vision Algorithms
<p>This repository makes available the dataset related to the paper:</p> <p>Zafer Çınar, Alessandro Russo, Matteo Spanio, Niccolò Pretto, and Sergio Canazza, <em>Filming the Sound: Anomaly Detection on Audio Tape Recordings using Computer Vision Algorithms</em>, IAI4CH, Bozen, 2024.</p> <p>The dataset and the experiment are described in the publication above.</p> <p>This repository contains two main directories (<strong>bold</strong> indicates directory names):</p> <ul> <li><strong>video samples</strong>: the actual videos used in the paper's experiment. This folder contains four subdirectories - 3.75 ips, 7.5 ips, 15 ips, and 30 ips - each representing a different playback speed (in inches per second). Within each subdirectory are several MP4 files, recorded on an A810 Studer open reel recorder, documenting the playback of magnetic audio tapes. The files follow the naming convention “Xips (Y).mp4,” where <em>X</em> represents the tape playback speed and <em>Y</em> is a serial number identifier for each video.</li> <li><strong>irregularities</strong>: the metadata for each video with timestamp and type of irregularity. The folder includes four CSV files - 3.75.csv, 7.5.csv, 15.csv, and 30.csv - corresponding to the playback speeds of the video samples. Each CSV file provides handmade annotations for its respective videos, with three columns: <ul> <li><em>video_id</em>: name of the video file in the format “Xips (Y).mp4,” where <em>X</em> is the tape speed and <em>Y</em> is the ID number.</li> <li><em>time_label</em>: timestamp indicating the irregularity, formatted as HH:MM:SS.mls.</li> <li><em>irregularity_type</em>: category of the detected anomaly, which may be one of the following: “splice,” “shadow,” “end-of-tape,” or “annotation.”</li> </ul> </li> </ul>
Image Data Collection Using a Multi-spectral Camera for Computer Vision Algorithms Research and Development
ClinicalTrials.gov study NCT05674149. IPD Sharing: NO. Countries: 1. Publications: 0.
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.