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.
12
datasets available to search
ShareScore release 0.9.0
Dataset results
12 results for “Fruit Detection”
AGS_apple_detection - Apple fruit images dataset for full image object detection
<p>This dataset correspond to full apple tree images (623) annotated for the task of object detection with its corresponding annotations in yolo format saved as txt files. The dataset was divided into test, train and validation<br><br>The data was collected in 2017 on 4 different apple varieties using a Samsung sm-a510F cell phone at two different resolutions: 2448 x 3264 px and 3096 x 4128 px in the orchards of Agroscope located in Wallis, Switzerland. </p>
Figure 1 in Detection of Male Mediterranean Fruit Flies (Diptera: Tephritidae): Performance of Trimedlure Relative to Capilure and Enriched Ginger Root Oil
Figure 1. Capture of C. capitata males in TML- versus CPL-baited Jackson traps for 3 replicates in an Oahu coffee field. Abscissa represents period of lure ageing, where 0 weeks represents fresh lures. Bar heights indicate mean of 20 traps per lure type; error bars represent + 1 SE. Symbols above bars show results of the Tukey HSD test comparing the 2 lures for each ageing category, where an asterisk indicates P <0.001 and ns indicates no significant difference.
deepNIR: Dataset for generating synthetic NIR images and improved fruit detection system using deep learning techniques
<p>In this paper, we present datasets that can be utilised for synthetic near infrared (NIR) image and bounding box level fruit detection system. It is undeniable fact that high-caliber machine learning software frameworks such as Tensorflow or Pytorch and large scale dataset such as ImageNet and COCO, and accelerated GPU hardware support have pushed the limit of machine learning for more than decades.</p> <p>Among these breakthroughs quality dataset is one of important key building blocks that can lead to success in model generalisation and deployment for data-driven deep neural networks. Particularly, synthetic data generation such as generative adversarial networks often requires relatively larger scale data than other supervised approaches. In addition, posing constrains such as geometrical facial constrains in fake face generation or consistent and radiometrically calibrated reflectances from satellite imagery commonly yield better results. We share NIR+RGB dataset that are re-processed from other two public datasets (nirscene and SEN12MS) and our own novel sweetpepper dataset to be able to timely adopt to other following studies.</p> <p>We oversampled from original nirscene dataset at 10, 100, 200, and 400 ratios and total of 127k pair of images. For SEN12MS satellite multispectral dataset, we selected one largest subset; Summer (45k) and All seasons (180k). Our sweetpeppr dataset consists of 1,615 pairs of NIR+RGB images. We demonstrate these NIR+RGB datasets are sufficient to be used for synthetic NIR generation quantitatively and qualitatively. We achieved Frechet Inception Distance (FID) of 11.36, 26.53, and 40.15 for nirscene1, SEN12MS, and sweetpepper dataset respectively.</p> <p>We also release <em>11</em> fruits' bounding box annotations that can be exported as various formats using cloud service. 4 newly added fruits [blueberry, cherry, kiwi, and wheat] compounds 11 novel bounding box dastaset together with our previous work in deepFruits project [apple, avocado, capsicum, mango, orange, rockmelon, strawberry]. The total number of bounding box instances is 162k and all bounding box dataset is ready for use from cloud service. For evaluation of these dataset, Yolov5 single stage detector is exploited and reported impressive mean-average-precision, mAP[0.5:0.95] results of [min:0.49, max:0.812]. We hope these dataset is useful and serves as one of baseline for the following up studies.</p>
Data to How to best detect threatened deadwood fungi – comparing metabarcoding and fruit body surveys
<p>This data set contains occurence data of threatened (red listed) fungi on deadwood objects in the National Park Bavarian Forest and the Žofínský prales National Nature Reserve to the the publication "How to best detect threatened deadwood fungi – comparing metabarcoding and fruit body surveys" by Rieker et al. (2024). This data is mainly uploaded to recreate analyses. </p> <p>Site: Information on the sampling site</p> <p>ObjectID: ID of the sampled deadwood object</p> <p>Campaign: Sampling year</p> <p>Decaystage: Decaystage of the deadwood object at sampling time (ranging between 1 and 4)</p> <p>Kingdom: Kingdom the eadwood object tree species belongs to</p> <p>Method: sampling method</p> <p>Species: threatend species name </p> <p>Presence: only contains the value "1" for present</p>
Figure 4 in Wafers in Saddle Bags: A Novel Dispensing System for Male Lures Used to Detect Invasive Fruit Flies (Diptera: Tephritidae)
Figure 4. Captures of Ceratitis capitata males in Jackson traps baited with either a polymeric plug containing 2 g TML or a saddle-bag containing 6 g TML at an Oahu (A) coffee field (wild males) or (B) citrus orchard (released males). Symbols represent means + 1 SE, where n = 15 traps per treatment per weathering interval at the coffee field and n = 12 at the citrus orchard. For a given weathering interval, means marked by different letters were significantly different (P <0.05, Holm-Šídák test).
Figure 3 in Wafers in Saddle Bags: A Novel Dispensing System for Male Lures Used to Detect Invasive Fruit Flies (Diptera: Tephritidae)
Figure 3. Captures of Bactrocera dorsalis males in Jackson traps baited with either a cotton wick containing 6 mL ME or a saddle-bag containing 6 g ME at study sites on Hawaii island or Oahu. Both fresh and aged wicks were deployed on Hawaii, but only fresh wicks were deployed on Oahu. Symbols represent means + 1 SE, where n = 15 traps per treatment per weathering interval for both study sites. For a given weathering interval, means marked by different letters were significantly different (P <0.05, Holm-Šídák test).
Figure 2 in Wafers in Saddle Bags: A Novel Dispensing System for Male Lures Used to Detect Invasive Fruit Flies (Diptera: Tephritidae)
Figure 2. Saddle bag dispensers. Top row (l to r): ME saddle bag in hand, with DDVP saddle bag on hanger; placement of ME saddle bag over DDVP saddle bag; hanger positioned inside Jackson trap. Bottom row (l to r): TML saddle bag in hand; TML saddle bag placed on hanger; hanger positioned inside Jackson trap.
Figure 1 in Wafers in Saddle Bags: A Novel Dispensing System for Male Lures Used to Detect Invasive Fruit Flies (Diptera: Tephritidae)
Figure 1. Standard method of baiting Jackson traps. Top row: Cotton wick containing ME in perforated basket and basket positioned inside Jackson trap. Bottom row: Polymeric plug containing TML in perforated basket and basket positioned inside Jackson trap.
Sainfoin Fruit Processing - Object Detection Dataset
<p>This dataset consists of 500 images of sainfoin (Onobrychis viciifolia) seed pods, seed, and split seeds. The images were taken as a part of an experiment to determine minimum sample size of seed pods needed to accurately estimate pod threshing trait heritability within sainfoin breeding lines.</p><p>The experiment was a complete factorial design with the following factors:</p><ul><li>Sainfoin named varieties: AAC Mountainview, Delaney, Eski , Rocky Mountain Remont, and Shoshone</li><li>Sample Size: 1, 2, 3, 4, and 5 grams of dried seed pods</li><li>Two different threshing types: Belt thresher processed 3X, Haldrup Impact Thresher (35sec @ Speed 9)</li></ul><p>This makes for a total factorial combination set of 5 varieties X 5 sample sizes X 2 threshing types = 50.</p><p>Each combination was comprised of 10 individual replicates where each replicate in a combination was a unique, random sample of seeds of the same mass (So, 10 random, 2g samples of Eski seed, processed by belt thresher; 10 random, 5g samples of Delaney seed processed by the Haldrup thresher, etc.). This makes for a total of 500 experimental units that comprise the sample set.</p><p>Once the seeds were sampled, weighed, and processed through the threshing equipment, they were weighed again and imaged.</p><p>The threshed seeds were scattered onto an imaging platform with a blue background, lit by 2 LED panels, and photographed with a Sony ILCE-7RM2 at the following settings:</p><ul><li>ISO: 100</li><li>Exposure: 1/40s</li><li>Focal Length: 55mm</li><li>Format: TIFF</li><li>Size: 7968x5320</li></ul><p>The raw images were converted from TIFF files to JPEG format and annotated in image labeling software. The seed objects were annotated with bounding boxes classified as the following classes</p><ol><li>pod: an enclosed seed pod</li><li>seed: a seed which was successfully threshed from the legume pod carpel</li><li>split: a seed threshed from the pod, but which split in two halves during the threshing process</li></ol><p>All image annotations were exported into the convenient <a href="https://docs.aws.amazon.com/rekognition/latest/customlabels-dg/md-coco-overview.html">COCO format</a>.</p><p>No further image processing was performed.</p><p>The image set was split into a 80/20 training and validation step using `scikit-learn` in Python 3.11 stratifying the datasets equally over the various experimental factor levels.</p><p>The zip file 'train_val_images.zip' contains a 'train' folder with 400 training images, 'val' containing 100 validation images, an image taken with a color correction card named 'color_test.jpg', and a json file with all the annotations.</p><p>Another file called 'seed_weights.csv' contains the image_name to global-key mapping in tabular format as well as the before and after threshing seed weights for each experimental sample.</p><p><strong>Labeling Metrics:</strong></p><ul><li>Pod (48.58%)<ul><li>36,599 objects</li></ul></li><li>Seed (33.83%)<ul><li>25,488 object</li></ul></li><li>Split (17.59%)<ul><li>13,255 objects</li></ul></li><li><strong>TOTAL (100%)</strong><ul><li><strong>75,342 objects</strong></li></ul></li></ul>
Date Fruit Detection Dataset for Computer Vision-Based Automatic Harvesting.
<p>The "Date Fruit Detection Dataset for Computer Vision-Based Automatic Harvesting" is a collection of videos and images showcasing date fruits from four Moroccan varieties, namely Majhoul, Boufagous, Bouisthami, and Khoult. This dataset is specifically designed to detect and classify date fruits, with the primary goal being the automation of the harvesting process.</p><p>All the images in this dataset were captured in two orchards located in Morocco, with the first orchard situated in the southeast of Errachidia and the second in Tismoumine, Alnif, Tinghir. These images were taken under various natural conditions, encompassing differing lighting, contrast, shadows, and instances where the dates were concealed by bags or hidden beneath palm leaves.</p><p>The dataset was meticulously compiled over the period spanning from June to September 2022, ensuring comprehensive coverage of all four maturity stages of date fruits, which include immature, khalal, rutab, and tamer. The dataset is intended for both object detection and classification purposes, and it includes a YOLO annotation txt file for each image. These annotations have been tailored to precisely recognize not only the date fruit but also to distinguish the specific variety and its maturity stage.</p>
Figure 2 in Detection of Male Mediterranean Fruit Flies (Diptera: Tephritidae): Performance of Trimedlure Relative to Capilure and Enriched Ginger Root Oil
Figure 2. Capture of C. capitata males in TML- versus EGRO-baited Jackson traps for a single replicate in an Oahu coffee field. Abscissa represents period of lure ageing, where 0 weeks represents fresh lures. Bar heights indicate mean of 20 traps per lure type; error bars represent + 1 SE. Symbols above bars show results of the Tukey HSD test comparing the 2 lures for each ageing category, where an asterisk indicates P <0.001 and ns indicates no significant difference.
DeepFruits: A Fruit Detection System Using Deep Neural Networks
<p>This is the dataset associated with MDPI Sensors paper entitled "DeepFruits: A Fruit Detection System Using Deep Neural Networks".</p> <p>This paper presents a novel approach to fruit detection using deep convolutional neural networks. The aim is to build an accurate, fast and reliable fruit detection system, which is a vital element of an autonomous agricultural robotic platform; it is a key element for fruit yield estimation and automated harvesting. Recent work in deep neural networks has led to the development of a state-of-the-art object detector termed Faster Region-based CNN (Faster R-CNN). We adapt this model, through transfer learning, for the task of fruit detection using imagery obtained from two modalities: colour (RGB) and Near-Infrared (NIR). Early and late fusion methods are explored for combining the multi-modal (RGB and NIR) information. This leads to a novel multi-modal Faster R-CNN model, which achieves state-of-the-art results compared to prior work with the F1 score, which takes into account both precision and recall performances improving from 0.807 to 0.838 for the detection of sweet pepper. In addition to improved accuracy, this approach is also much quicker to deploy for new fruits, as it requires bounding box annotation rather than pixel-level annotation (annotating bounding boxes is approximately an order of magnitude quicker to perform). The model is retrained to perform the detection of seven fruits, with the entire process taking four hours to annotate and train the new model per fruit.</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.