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.
979
datasets available to search
ShareScore release 0.9.0
Dataset results
979 results for “image dataset”
Synethic dataset for differential OGI images and wind tunnel dataset
<h2>Fluid Flow Dataset</h2> <p> </p> <h2>1. Introduction</h2> <p>A total of three datasets are included here, namely, (1) the synthetic dataset used for model training as mentioned in the paper, (2) the wind tunnel dataset used to validate the segmentation performance of the model, including the corresponding manually labeled labels, and (3) the original wind tunnel experiment dataset.</p> <h2>2. Synthetic differential image dataset for segmentation</h2> <p>To address the lack of semantic segmentation datasets for infrared fluid flow imagery, we created a synthetic dataset from the ScalarFlow dataset, focusing on pixel-level labels suitable for neural network training. Our process, illustrated as shown in the paper, includes generating realistic noise by capturing images under controlled conditions with an Optical Gas Imaging (OGI) camera, followed by image subtraction, normalization, and merging with ScalarFlow data. This method ensures the inclusion of real-world disturbances such as camera jitter effects, enhancing the dataset's robustness and applicability. The dataset, enriched with various data augmentation techniques, comprises over 30,000 images split into training, validation, and testing sets, catering to the rigorous demands of practical applications in fluid dynamics analysis.</p> <h2>3. Wind tunnel dataset</h2> <p>To enable the determination of velocities of fluid flow by using optical flow algorithms, a wind tunnel data set that includes fluid images captured by different cameras was recorded. </p> <p>The fluid flow is created in the wind tunnel and generated by different substances, i.e., dry ice, smoke matches, or paraffin oil. In addition, velocity data collected from the 3D ultrasonic anemometer were used as a reference to evaluate the performance and accuracy of the optical flow algorithms. The fluid flow rate was set at three different velocities in Euclidean space, i.e., 0.7 m/s, 1.4 m/s, and 2.0 m/s.</p> <div> <div>This dataset is captured by using a wind tunnel, the OGI camera FLIR GF320 and a 3D anemometer for obtaining reference flow velocities. Below is the information of the used camera.</div> </div> <h3>FLIR GF320 Camera Info</h3> <table> <tbody> <tr> <td> <div> <div><strong>Parameter</strong></div> </div> </td> <td> <div> <div><strong>Value</strong></div> </div> </td> </tr> <tr> <td> <div> <div>Spectral Range</div> </div> </td> <td> <div> <div>3.2 – 3.4 μm</div> </div> </td> </tr> <tr> <td> <div> <div>Standard Temperature Range</div> </div> </td> <td> <div> <div>–20°C to +350°C</div> </div> </td> </tr> <tr> <td> <div> <div>Accuracy</div> </div> </td> <td> <div> <div> ±1 °C for 0 °C to 100 °C; ±2% > 100 °C</div> </div> </td> </tr> <tr> <td> <div> <div>Lenses</div> </div> </td> <td> <div> <div>24° × 18°</div> </div> </td> </tr> <tr> <td> <div> <div>Resolution</div> </div> </td> <td> <div> <div>320 × 240 Pixel</div> </div> </td> </tr> </tbody> </table> <p> </p> <p>The dataset consists of three parts, i.e., OGI images of fluids/smoke generated from three different substances.</p> <ul> <li>1. Smoke matches dataset<br> <br> <table> <tbody> <tr> <td> <div> <div><strong>Id</strong></div> </div> </td> <td><strong>Substances</strong></td> <td> <div> <div><strong>Velocity in m/s</strong></div> </div> </td> <td> <div> <div><strong>Group</strong></div> </div> </td> <td> <div> <div><strong>Number of frames</strong></div> </div> </td> </tr> <tr> <td>1</td> <td>Smoke matches</td> <td>0.7</td> <td>1</td> <td>143</td> </tr> <tr> <td>2</td> <td>Smoke matches</td> <td>0.7</td> <td>2</td> <td>597</td> </tr> <tr> <td>3</td> <td>Smoke matches</td> <td>1.4</td> <td>1</td> <td>145</td> </tr> <tr> <td>4</td> <td>Smoke matches</td> <td>1.4</td> <td>2</td> <td>598</td> </tr> <tr> <td>5</td> <td>Smoke matches</td> <td>2.0</td> <td>1</td> <td>145</td> </tr> <tr> <td>6</td> <td>Smoke matches</td> <td>2.0</td> <td>2</td> <td>596</td> </tr> </tbody> </table> </li> <li>2. Paraffin oil dataset<br><br> <table> <tbody> <tr> <td> <div> <div><strong>Id</strong></div> </div> </td> <td><strong>Substances</strong></td> <td> <div> <div><strong>Velocity in m/s</strong></div> </div> </td> <td> <div> <div><strong>Group</strong></div> </div> </td> <td> <div> <div><strong>Number of frames</strong></div> </div> </td> </tr> <tr> <td>7</td> <td> <div> <div>Paraffin oil</div> </div> </td> <td>0.7</td> <td>1</td> <td>597</td> </tr> <tr> <td>8</td> <td>Paraffin oil</td> <td>0.7</td> <td>2</td> <td>597</td> </tr> <tr> <td>9</td> <td>Paraffin oil</td> <td>1.4</td> <td>1</td> <td>597</td> </tr> <tr> <td>10</td> <td>Paraffin oil</td> <td>1.4</td> <td>2</td> <td>597</td> </tr> <tr> <td>11</td> <td>Paraffin oil</td> <td>2.0</td> <td>1</td> <td>597</td> </tr> <tr> <td>12</td> <td>Paraffin oil</td> <td>2.0</td> <td>2</td> <td>597</td> </tr> </tbody> </table> </li> <li>3. Dry ice dataset<br><br> <table> <tbody> <tr> <td> <div> <div><strong>Id</strong></div> </div> </td> <td><strong>Substances</strong></td> <td> <div> <div><strong>Velocity in m/s</strong></div> </div> </td> <td> <div> <div><strong>Group</strong></div> </div> </td> <td> <div> <div><strong>Number of frames</strong></div> </div> </td> </tr> <tr> <td>13</td> <td>Dry ice</td> <td>0.7</td> <td>1</td> <td>598</td> </tr> <tr> <td>14</td> <td>Dry ice</td> <td>0.7</td> <td>2</td> <td>597</td> </tr> <tr> <td>15</td> <td>Dry ice</td> <td>1.4</td> <td>1</td> <td>595</td> </tr> <tr> <td>16</td> <td>Dry ice</td> <td>1.4</td> <td>2</td> <td>596</td> </tr> <tr> <td>17</td> <td>Dry ice</td> <td>2.0</td> <td>1</td> <td>185</td> </tr> <tr> <td>18</td> <td>Dry ice</td> <td>2.0</td> <td>2</td> <td>628</td> </tr> </tbody> </table> </li> </ul> <p> </p> <p>We also provide the corresponding 3D anemometer data, which allows the user to convert the pixel displacement from the image to the actual flow rate, as shown in below.</p> <div> <h3>Velocities in m/s and pixel</h3> <table> <tbody> <tr> <td> <div> <div><strong>Velocity in m/s</strong></div> </div> </td> <td> <div> <div><strong>Settings of WindChannel</strong></div> </div> </td> <td> <div> <div><strong> Velocity in pixel</strong></div> </div> </td> </tr> <tr> <td> <div> <div>0.7</div> </div> </td> <td>1.88</td> <td>4.57</td> </tr> <tr> <td>1.4</td> <td>2.50</td> <td>9.14</td> </tr> <tr> <td>2.0</td> <td>3.06</td> <td>13.06</td> </tr> </tbody> </table> <p> </p> <h2>4. Wind tunnel segmentation dataset</h2> <p>This part of the dataset is from the dry ice dataset portion of the wind tunnel test dataset described above. And labels are generated by manual labeling for evaluating the performance of the image segmentation model in real-world scenarios, a total of 100 differential images and 100 labels.</p> <p> </p> </div>
Citrus Tree Root System Image Dataset: Effects of Propagation Methods on Root Architecture
<p>This dataset is composed of images from two different citrus rootstock trials designated "Field Trial 1" and "Field Trial 2". Each field trial was planted with trees of <em>Citrus sinensis</em> cv. 'Valencia' grafted onto 4 different, commercially available USDA citrus rootstocks: US-812, US-897, US-942, US-1516. The trees were excavated two years after planting, cleaned, dried, and imaged in two ways. First, the root systems were imaged radially as though you are looking down through the trunk of the tree with the roots radiating outward in all direction. Secondly, the root systems were imaged vertically in six different positions such that each image was a side view of the root system from a different angle. Additionally, the binary masks of the vertical root systems were included in each .tar file.</p> <p>All images were acquired with a Canon EOS Rebel T6 against a professional photography blue screen. A mapping of the image names to the experimental trial information is included in each .tar file, and a series of scaled images were taken before and after root system imaging to calculate the pixel to centimeter conversion for absolute measurements.</p> <p>This dataset is currently in beta as it may change at some point in the future, therefore until that time, this record will be set to restricted.</p>
A comprehensive dataset of magnetic resonance enterography images with bowel segment annotations
<p>Inflammatory bowel disease (IBD) is a recurrent bowel disease that usually requires magnetic resonance enterography (MRE) for diagnosis and monitoring. However, recognition of bowel segments from MRE images by a radiologist is challenging and time-consuming. Deep learning-based medical image segmentation has shown the potential to reduce manual effort and provide automated tools to assist in disease management; however, it requires a large-scale fine<span>-</span>annotated dataset for training. To address this gap, we collected MRE data, including HASTE(half-Fourier acquisition single-shot turbo spin-echo) sequences with coronal orientation, from 114 patients with IBD. The bowel images per patient were contoured and annotated into ten segments (stomach, duodenum, small intestine, appendix, cecum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum), with fine pixel-level annotations labeled by experienced radiologists. Furthermore, we <span>validated</span> the efficiency of several state-of-the-art segmentation methods using this dataset. This study established a high-quality, publicly available whole-bowel segment MR dataset with benchmark results and laid the groundwork for AI research on IBD.</p>
STS-Tooth: A multi-modal dental dataset for semi-supervised deep learning image segmentation
<p>In response to the increasing prevalence of dental diseases, dental health, a vital aspect of human well-being, warrants greater attention. Panoramic X-ray images (PXI) and Cone Beam Computed Tomography (CBCT) are key tools for dentists in diagnosing and treating dental conditions. Additionally, deep learning for tooth segmentation can focus on relevant treatment information and localize lesions. However, the scarcity of publicly available PXI and CBCT datasets hampers their use in tooth segmentation tasks. Therefore, this paper presents a multimodal dataset for semi-supervised deep learning in dental PXI and CBCT, named STS-2D-Tooth and STS-3D-Tooth. STS-2D-Tooth includes 4,000 images and 900 masks, categorized by age into children and adults. Moreover, we have collected CBCTs providing more detailed and three-dimensional information, resulting in the STS-3D-Tooth dataset comprising 148,400 unlabeled scans and 8,800 masks. To our knowledge, this is the first multimodal dataset combining dental PXI and CBCT, and it is the largest tooth segmentation dataset, a significant step forward for the advancement of tooth segmentation.</p>
NasalSeg Dataset for Nasal Cavity and Paranasal Sinuses Segmentation from CT Images
<p>NasalSeg is the first large-scale, open-access dataset for nasal cavity and paranasal sinus segmentation from 3D CT images. The dataset comprises 130 CT scans with pixel-wise annotation of five anatomical structures including the left nasal cavity, right nasal cavity, nasal pharynx, left maxillary sinus, and right maxillary sinu.</p>
Thyroid scan image dataset for the study of thyroid pathologies in adult patients
<p>The Dataset containing 641 images of Thyroid Gammagraphies studies corresponding to 235 patients over 18 years of age that were acquired in the period from 2016 to 2024 at the Instituto de Investigaciones en Ciencias de la Salud, Universidad Nacional de Asunción (UNA), Paraguay. The thyroid gammagraphies images were acquired on the trimodal SPECT-CT-PET equipment, model AnyScan SCP, MEDISO brand.</p> <p>The dataset contains 16 folders, corresponding to the different types of diagnosed pathologies. Within each folder the images were grouped again into 3 folders according to the projections used to acquire the images: Anterior, RAO and LAO. This new dataset includes 641 images for each type of projection, all related to diagnoses of thyroid pathologies obtained from thyroid scans. The labels of each image, along with its respective diagnosis, are detailed in the file called <em>Classification Annotations.xlsx</em></p> <p>The images were classified by the professionals of the Nuclear Medicine Service, according to the diagnoses made by the nuclear physicians and were grouped into:</p> <ul> <li>Toxic adenoma.</li> <li>Diffuse goiter.</li> <li>Multinodular goiter.</li> <li>Nodular goiter.</li> <li>Absent thyroid gland - Total thyroidectomy.</li> <li>Preserved thyroid gland.</li> <li>Deformed thyroid gland.</li> <li>Right hemithyroidectomy.</li> <li>Autonomous nodule.</li> <li>Hyperuptake nodule.</li> <li>Hypouptake nodule.</li> <li>Remnant after Total thyroidectomy.</li> <li>Iatrogenically blocked thyroid.</li> <li>De-Quervain's subacute thyroiditis.</li> <li>Diffuse goiter - Subacute thyroiditis.</li> <li>Multinodular goiter - Subacute thyroiditis.</li> </ul>
Dataset for "Imaging the ion-molecule reaction dynamics of O- + CD4"
<p>Dataset for "Imaging the ion-molecule reaction dynamics of O- + CD4"</p>
Astronomical Image Classification Dataset
<p>Dataset for the work published in SPIE Sensors and Imaging 2023:</p> <p>Keenan A. A. Chatar, <a href="https://www.spiedigitallibrary.org/profile/ezrafielding">Ezra Fielding</a>, <a href="https://www.spiedigitallibrary.org/profile/Kei.Sano-4227702">Kei Sano</a>, and <a href="https://www.spiedigitallibrary.org/profile/Kentaro.Kitamura-4492438">Kentaro Kitamura</a> "Data downlink prioritization using image classification on-board a 6U CubeSat", Proc. SPIE 12729, Sensors, Systems, and Next-Generation Satellites XXVII, 127290K (19 October 2023); <a href="https://doi.org/10.1117/12.2684047" target="_blank" rel="noopener">https://doi.org/10.1117/12.2684047</a></p>
Dataset for: Country Image in Latin America: A Systematic Review of Research in Scopus, Web of Science, and SciELO
<p>Dataset for: Country Image in Latin America: A Systematic Review of Research in Scopus, Web of Science, and SciELO</p>
Dataset supporting the paper "Characterization of a mouse-sized receive-only coil for a preclinical magnetic particle imaging scanner"
<p>We have recently submitted a paper on the characterization of a mouse-sized receive-only coil for the Bruker preclinical MPI scanner. This dataset contains the measurements for our comparison between the native transmit-receive (TxRx) coil of the scanner and the new mouse-sized receive-only (Rx) coil. </p> <p>Included are: </p> <ul> <li>Network analyzer (NA) measurements as csv, first column=frequency, second column=power transfer. There are measurements at the center of the coils and along x- and y-axis with a 2mm step size.</li> <li>System matrix (SM) measurements, including the background (BG) measurements and the calculated particle signal-to-noise ratio (SNR) (Bruker file format).</li> <li>MPI measurements of a dilution series (Bruker file format). In the dilution series folders there is a readme, which sample "#A" was measured at which position "_A"/"_B"/"_C", is saved in which folder "(E#)" and contains which iron amount (in ng). </li> </ul>
2D image dataset for the presence or absence of aesthetical defects on EV battery welds.
<p>A 2D image dataset of EV car battery welds is used to train AI/ML models on weld classification for quality inspection applications or other purposes.</p> <p> </p> <p>Includes 1 zip file with 61,670 greyscale images of 409x40 resolution split in</p> <ul> <li>training set [ 43,169 images]</li> <li>validation set [ 9,250 images]</li> <li>testing set [ 9,251 images]</li> </ul> <p> </p> <p>and the following 2 classes:</p> <p>1] OK --> indicating the absence of aesthetical defects</p> <p>2] KO --> indicating the presense of aesthetical defects</p>
Dataset for "Fiber-Seismometer Hybrid Sensing for Seismic Imaging and Monitoring"
<p>The ambient noise dataset was collected on May 8, 2023, from a fiber-seismometer hybrid sensing deployment positioned along the Qiantang River in Hangzhou. The DAS data, Z-component and R-component data of seismometers are all stored in MAT format. Please refer to our study for detailed information on the dataset.</p> <p>Abstract about this study:</p> <p>Extreme climate events and geological disasters have intensified the urgency for advancing seismic imaging and monitoring. Despite developments in seismic instrumentation, particularly with seismometers and Distributed Acoustic Sensing (DAS), fine-scale observations remain challenging due to their inherent limitations and deployment configurations. This study introduces a novel hybrid sensing interferometry method that enhances multi-component signal extraction—especially poor horizontal components—through a two-step cross-correlation of DAS and seismometers. A field application near the Qiantang River in Hangzhou illustrates how our proposed framework retrieves high-quality multi-component empirical Green’s functions and advances ultra-short duration ambient noise seismic imaging techniques, including surface wave dispersion measurements and horizontal-to-vertical spectral ratio assessments. Our approach also facilitates monitoring of near-surface seismic velocity changes, dv/v, with an unprecedented 10-minute resolution, shedding light on shallow dynamic hydraulic responses. This innovative hybrid sensing framework offers new perspectives and methodologies for transforming future research in seismological observation, imaging, and monitoring. </p>
Bone Scan Images Dataset for Study of Bone Metastases in Adult Breast Cancer Patients at IICS-UNA Paraguay
<p>This dataset contains 582 bone scan images from 291 adult patients who attended the Nuclear Medicine Service at the Instituto de Investigaciones en Ciencias de la Salud (IICS) of the Universidad Nacional de Asunción (UNA), Paraguay, between 2020 and 2024.</p> <p>The dataset is organized into two folders, each named according to the classification of the images: bone metastases and no bone metastases. Each of these is further subdivided into folders corresponding to the projections generated during the acquisition of the bone scan images, anterior and posterior.</p>
Example dataset for in vivo blood sO2 imaging with ultrasound-aided large-scale optoacoustic microscopy
<p>"mouse1_r_5um_dualwave.mat" and "mouse1_r_10um_us.mat" contain the raw datasets in dual-wavelength optoacoustic and ultrasound modes acquired with ultrasound-aided large-scale optoacoustic microscopy on mouse dorsal skin.</p> <p>"mouse1_r.mat" contains the estimated skin surface based on the ultrasound dataset.</p> <p>After downloading the above 3 files, make a subfolder under the current working folder and name it "skinCropDepths", and put "mouse1_r.mat" under that subfolder.</p> <p>The code can be fetched from this Github repository: https://github.com/razanskylab/dualwaveProcessor. Refer to README file for explanation on the code base.</p> <p>The datasets shared here were acquired on healthy mouse dorsal skin and are intended for testing the unmixing scripts.</p> <p>The longitudinal datasets acquired during dorsal skin wound healing are too large to be publically shared, but is available for research purposes upon reasonable request: weiye.li@uzh.ch.</p> <p> </p> <p> </p>
Dataset for Crack Detection in Images of Masonry Using CNNs
<p>We trained a convolutional neural network (CNN) on images of brick walls built in a laboratory environment and test its ability to detect cracks in images of brick-and-mortar structures both in the laboratory and on real-world images taken from the internet. We also compared the performance of the CNN to a variety of simple classifiers operating on handcrafted features. This is the dataset used in that work.</p>
Convallaria dataset for microscopy image denoising benchmark as used in Probabilistic Noise2Void paper
<p>Convallaria dataset for microscopy image denoising benchmark as used in Probabilistic Noise2Void paper (https://ieeexplore.ieee.org/document/9098336)</p>
Subset of BioID dataset (https://www.bioid.com/facedb/) used for image denoising benchmark as used in DivNoising paper (https://arxiv.org/abs/2006.06072)
<p>The original BioID dataset comes from https://www.bioid.com/facedb/. </p> <p>A subset of original BioID dataset was used for image denoising benchmark (corrupted with zero mean Gaussian noise of std 15) as in DivNoising paper (https://arxiv.org/abs/2006.06072)</p>
Flywing (noise 10) dataset for microscopy image denoising and segmentation benchmark as used in DenoiSeg paper (https://arxiv.org/abs/2005.02987)
<p>Flywing n10 dataset for microscopy image denoising and segmentation benchmark as used in DenoiSeg paper (https://arxiv.org/abs/2005.02987)</p>
Flywing (noise 0) dataset for microscopy image denoising and segmentation benchmark as used in DenoiSeg paper (https://arxiv.org/abs/2005.02987)
<p>Flywing n0 dataset for microscopy image denoising and segmentation benchmark as used in DenoiSeg paper (https://arxiv.org/abs/2005.02987)</p>
DSB (noise 20) dataset for microscopy image denoising and segmentation benchmark as used in DenoiSeg paper (https://arxiv.org/abs/2005.02987)
<p>DSB n20 dataset for microscopy image denoising and segmentation benchmark as used in DenoiSeg paper (https://arxiv.org/abs/2005.02987)</p>
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.