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.
505
datasets available to search
ShareScore release 0.9.0
Dataset results
505 results for “apples”
FIGURES 27–31. Ologamasus tuberculatus n in Two new species of the genus Ologamasus (Ologamasidae) from apple orchards in southern Brazil
FIGURES 27–31. Ologamasus tuberculatus n. sp. Adult female. 27. Dorsal idiosoma; 28. Ventral idiosoma; 29. Epistome; 30. Chelicera; 31. Palp.
FIGURES 18–22. Ologamasus tuberculatus n in Two new species of the genus Ologamasus (Ologamasidae) from apple orchards in southern Brazil
FIGURES 18–22. Ologamasus tuberculatus n. sp. Protonymph. 18. Dorsal idiosoma; 19. Ventral idiosoma; 20. Epistome; 21. Palp genu; 22. Chelicera.
FIGURES 3–6. Ologamasus margaridae n in Two new species of the genus Ologamasus (Ologamasidae) from apple orchards in southern Brazil
FIGURES 3–6. Ologamasus margaridae n. sp. Deutonymph. 3. Dorsal idiosoma; 4. Ventral idiosoma; 5. Tritosternum; 6. Chelicera.
Dataset for the paper "Nanostructured Catalyst Layer Allowing Production of Ultralow Loading Electrodes for Polymer Electrolyte Membrane Fuel Cells with Superior Performance" published in ACS Appl. Energy Mater.
<p>The data in this spreadsheet was used to produce the figures in the paper</p><p>Authors:</p><p>Colleen Jackson, Michalis Metaxas, Jack Dawson, Anthony Kucernak</p><p>Title:</p><p>Nanostructured Catalyst Layer Allowing Production of Ultralow Loading Electrodes for Polymer Electrolyte Membrane Fuel Cells with Superior Performance</p><p>Journal:</p><p>ACS Appl. Energy Mater. </p><p>DOI:</p><p>10.1021/acsaem.3c01987</p><p>Please cite the above reference if you wish to use this data</p><p>DOI of data:</p><p>10.5281/zenodo.10256698</p>
SpectroFood dataset Apple
<p>This dataset contains hyperspectral data in the visible / shortwave near-infrared spectral domain (430 - 990 nm) for apple. The data set contains cubes of 240 samples, wavelength array and sample labels in Matlab's mat format.</p> <p>The data set is coupled with dry matter content references that are provided here: <a href="../records/8362947">https://zenodo.org/records/8362947</a></p> <p> </p> <p> </p>
Personal Digital Assistant (PDA) Apple Newton
The Newton portable computer came out of the first results of Apple's work on a miniature mobile device with a touchscreen and operated with a stylus. The design offered a range of breakthrough solutions, one of them being the handwriting recognition system. The computer's design used a high performance, energy efficient CPU of the ARM family, which later became standard in this type of device and is still used in tablets and smartphones. The Newton's main purpose was time management and performing simple office tasks. The content of its built-in memory could be synchronised with computers using the Windows or MacOS operating systems over a serial interface. A wireless connection was also available in the form of an infrared port. An additional extension card allowed the Newton to be connected to the telephone network in order to send and receive faxes or use basic Internet functionalities. Apple Computer Inc., USA (design)/Japan (production), 1993 Inv. No. MIM1414/VII-134 Licence: CC BY-NC-SA Source: Objaverse 1.0 / Sketchfab
Apple detection dataset
<p><strong>1.</strong><strong>背景介绍</strong></p> <p>苹果检测是计算机视觉与图像处理领域中的一个重要问题,其应用包括农业自动化、食品质量检测和无人机水果采摘等。在这些应用中,准确地检测和识别苹果对于提高生产效率、保证食品质量以及优化农业管理都具有重要意义。苹果检测的挑战之一在于其外观和形状多样,受光照、阴影、颜色变化以及遮挡等因素的影响较大。另外,苹果通常生长在树上,背景复杂,可能包含树叶、树枝、其他水果、土壤等干扰物体,因此需要建立鲁棒的检测方法来应对这些复杂情况。近年来,深度学习技术的发展为苹果检测带来了新的希望。</p> <p>基于深度学习的目标检测算法,如YOLO (You Only Look Once)和Faster R-CNN等,在苹果检测中取得了显著的成果。这些算法能够针对复杂背景和不同形状的苹果进行准确的检测和定位,为后续的质量评估和自动化采摘提供了可靠的基础。除了算法技术的发展,苹果检测还涉及到大规模数据集的采集和标注工作。精心采集的数据集可以帮助算法更好地理解苹果的外观特征,并提高检测的准确性和鲁棒性。同时,合适的数据增强技术也能够有效地提升算法在复杂场景下的表现。</p> <p>总的来说,苹果检测是一个具有挑战性但又充满潜力的研究领域。随着计算机视觉和深度学习技术的不断进步,相信苹果检测技术将在农业生产和食品加工领域发挥越来越重要的作用,为农业现代化和智能化注入新的活力。</p> <p><strong>2.</strong><strong>数据集介绍</strong></p> <p>本研究采用LabellImg标注工具对原始图像中的苹果目标绘制外接矩形框以实现人工标注。设定苹果标签为apple,保存格式为PASCAL VOC。苹果图像数据集共2825张。</p> <p><strong>2.1</strong><strong>数据格式</strong></p> <p><span>苹果检测数据集包括训练集、验证集、测试集和苹果的标注数据。</span><span>苹果图像数据集共2825张。将数据集按照9:1:1的比例随机分为2197幅图像用来训练模型, 314幅图像用来验证模型,314幅图像用来测试模型。</span></p> <p><strong>2.2</strong><strong>文件列表</strong></p> <p>本数据集包含如下文件:</p> <p>1. Train:训练集</p> <p>2. Valid:验证集</p> <p>3. Test:测试集</p> <p><span>4. </span><span><span><span><span> </span></span></span><span>Data.yaml: </span>描述了数据集的相关信息,包括数据文件的路径、标签信息、数据格式</span></p> <p>注:使用解压工具解压可能会失败,请使用命令行进行解压操作,如:unzip xxx.zip</p> <p><strong>3.</strong><strong>数据集类型</strong></p> <p>制造、图像识别</p>
Phased T2T reference genome and pangenome reveal expanded resistance gene analogs in apple domestication
<p>Phased T2T reference genome and pangenome reveal expanded resistance gene analogs in apple domestication<br>The data contains two files, namely genome file and genome annotation file:<br>1. Genome:<br>This file contains 12 genomes for 5 species (cultivar), namely: <em>Malus domestica</em> cv. ‘Golden Delicious’ (GD), <em>Malus domestica</em> cv. ‘Gala’ (Gala), <em>Malus domestica</em> cv. ‘Honeycrisp’ (HC), <em>Malus domestica</em> cv. ‘HFTH1’ (HFTH1), <em>Malus baccata</em> (Mba), <em>Malus sieversii</em> (Msi), and <em>Malus sylvestris</em> (Msy)</p> <p>2. Genome annotation:<br>Corresponding to the genome file.<br>Note: Hap1 and hap2 represent two haplotypes of a species (cultivar), all data used for the construction of pan-genomes and comparative genomics analysis.</p>
3D Point Clouds of Trees and Apple Fruit Annotated with Thermal Data
<p>The data set captures four measurements during fruit growth: 06/28/2022 (15:00), 07/12/2022 (15:00), 09/01/2022 (15:00), 09/06/2022 (13:00)</p> <p>Additionally, diurnal courses are provided for three days: </p> <table> <tbody> <tr> <td> <p>Date</p> </td> <td> <p>Time</p> </td> </tr> <tr> <td> <p>09/21</p> </td> <td> <p>07:00, 08:00, 10:00, 12:00, 13:00, 18:00</p> </td> </tr> <tr> <td> <p>09/22</p> </td> <td> <p>07:00, 08:00, 10:00, 12:00, 13:00, 18:00</p> </td> </tr> <tr> <td> <p>10/05</p> </td> <td> <p>06:30, 07:00, 09:00, 10:00, 11:00, 16:00</p> </td> </tr> </tbody> </table> <p> </p> <p>The data set was measured in Blocks (A-D) of trees (T) on apple (A) fruit and is stored as compressed zip files, capturing raw, preprocessed, and manually recorded reference (ground truth) data.</p> <p>1. zip files entitled Raw_YYYY_MM_DD_Block[A-C]-[R, L] for seasonal data and Raw_YYYY_MM_DD_BlockD-[R, L]_Hour__:__ for diel data</p> <p>- raw data of LiDAR 3D point clouds - txt files</p> <p>- raw image data by thermal camera - txt files</p> <p>2. zip files entitled YYYY_MM_DD or DailyAcquisitions:</p> <p>- preprocessed (merged) sensor data of temperature-annotated 3D point clouds of canopies - csv files</p> <p>- preprocessed data, capturing manually segmented point clouds of temperature-annotated fruit - txt files</p> <p>3. Microsoft Excel files entitled References and Weather data:</p> <p>- raw data, representing reference data of fruit - xlsx file</p> <p>- raw data of weather conditions - xlsx file</p>
Apple detection code
<p><strong><span>1.</span></strong><strong><span>背景介绍</span></strong></p> <p>苹果检测是计算机视觉与图像处理领域中的一个重要问题<span>,</span>其应用包括农业自动化、食品质量检测和无人机水果采摘等。在这些应用中<span>,</span>准确地检测和识别苹果对于提高生产效率、保证食品质量以及优化农业管理都具有重要意义。苹果检测的挑战之一在于其外观和形状多样<span>,</span>受光照、阴影、颜色变化以及遮挡等因素的影响较大。另外<span>,</span>苹果通常生长在树上<span>,</span>背景复杂<span>,</span>可能包含树叶、树枝、其他水果、土壤等干扰物体<span>,</span>因此需要建立鲁棒的检测方法来应对这些复杂情况。近年来<span>,</span>深度学习技术的发展为苹果检测带来了新的希望。</p> <p><span> </span></p> <p>基于深度学习的目标检测算法<span>,</span>如<span>YOLO (You Only Look Once)</span>和<span>Faster R-CNN</span>等<span>,</span>在苹果检测中取得了显著的成果。这些算法能够针对复杂背景和不同形状的苹果进行准确的检测和定位<span>,</span>为后续的质量评估和自动化采摘提供了可靠的基础。除了算法技术的发展<span>,</span>苹果检测还涉及到大规模数据集的采集和标注工作。精心采集的数据集可以帮助算法更好地理解苹果的外观特征<span>,</span>并提高检测的准确性和鲁棒性。同时<span>,</span>合适的数据增强技术也能够有效地提升算法在复杂场景下的表现。</p> <p><span> </span></p> <p>总的来说<span>,</span>苹果检测是一个具有挑战性但又充满潜力的研究领域。随着计算机视觉和深度学习技术的不断进步<span>,</span>相信苹果检测技术将在农业生产和食品加工领域发挥越来越重要的作用<span>,</span>为农业现代化和智能化注入新的活力。</p> <p><strong><span>2.</span></strong><strong><span>代码介绍</span></strong></p> <p>本研究采用<span>Python</span>语言对<span>YOLOv8s</span>苹果检测系统进行设计改进。</p> <p><strong><span>2.1</span></strong><strong><span>文件列表</span></strong></p> <p><strong><span>1.github</span></strong></p> <p><span>ISSUE_TEMPLATE:</span>提供不同类型的问题报告模板,包括<span> bug-report,yml</span>、<span>config,yml!feature-requestym</span>和<span> question.yml</span>。这些模板帮助用户以结构化的方式报告错误、提出功能请求或提问。</p> <p><span>workfows:</span>包含多个工作流文件,如<span>ciym(</span>持续集成<span>)</span>、<span>cla.yml(</span>贡献者许可协议<span>)</span>、<span>codeqlym(</span>代码质量检查<span>)</span>、</p> <p><span>docker.ym(Docker</span>配置<span>)</span>、<span>greetngs.yml(</span>自动问候新贡献者<span>)</span>、<span>links,ymlkpublish.yml(</span>自动发布<span>)</span>、<span>stale.yml(</span>处理陈旧问题<span>)</span></p> <p><span>dependabot:yml(</span>自动依赖更新<span>)</span></p> <p>这些文件共同支持项目的自动化管理,包括代码质量保证、持续集成和部署、社区互动和依赖项维护。</p> <p><strong><span>2.docker</span></strong></p> <p><span>Dockerfile:</span>主要的<span>Docker</span>配置文件,用于构建项目的默认<span>Docker</span>镜像。</p> <p><span>Dockerfile-arm64:</span>针对<span>ARM64</span>架构的设备<span>(</span>如某些类型的服务器或高级嵌入式设备<span>)</span>定制的<span>Docker</span>配置<span>,</span></p> <p><span>Dockerfile-conda:</span>使用<span>Conda</span>包管理器配置环境的<span>Docker</span>配置文件<span>:</span></p> <p><span>Dockerfile-cpu: </span>为不支持<span>GPU</span>加速的环境配置的<span>Docker</span>配置文件。</p> <p><span>Dockerfile-jetson: </span>专为<span>NVIDlA Jetson</span>平台定制的<span>Docker</span>配置。</p> <p><span>Dockerfile-python: </span>可能是针对纯<span>Python</span>环境的简化<span>Docker</span>配置</p> <p><span>Dockerfile-runner </span>可能用于配置持续集成特续部署<span>(CI/CD)</span>运行环境的<span>Docker</span>配置<span>,</span></p> <p>这些配置文件是用来部署用的,用户可以根据自己的需要选择合适的环境来部署和运行项目</p> <p><strong><span>3.docs</span></strong></p> <p><span>docs</span>日录通常用于存放文档资料,包括多种语言的翻译。例如,此目录下有多个文件夹,每个文件夹代表一种语言<span>(</span>如<span>en</span>代表英语文档<span>)</span>。除此之外,还有几个重要的<span>Python</span>脚本和配置文件给大家说一下<span>:</span></p> <p><span>build_docs.py:-</span>个<span>Python</span>脚本,用于自动化构建和编译文档的过程<span>,mkdocs.yml:MkDocs</span>配置文件,用于指定文档网站的结构和设置。</p> <p>以<span>mkdocs_es.yml</span>为例,这是用于构建西班牙语文档的<span>MKDocs</span>配置文件。类似的,<span>mkdocs zh.yml</span>用于构建中文文档。</p> <p><strong><span>4.examples</span></strong></p> <p><span>YOLO8-CPP-nference:</span>包含<span>C++</span>语言实现的<span>YOLOV8</span>推理示例,内有<span>CMakeLists.bt(</span>用于项目构建的<span>CMake</span>配置文件<span>)inference.cpp</span>和<span>inferenceh(</span>推理相关的源代码和头文件<span>)</span>,<span>main.cpp(</span>主程序入口<span>)</span>以及<span>README.md(</span>使用说明<span>)</span></p> <p><span>YOLOv8-0NNXRuntime:</span>提供<span>Pvthon</span>语言与<span>ONNXRunime</span>结合使用的<span>YOLOV8</span>推理示例,其中<span>main.py</span>是主要的脚本文件<span>README.md</span>提供了如何使用该示例的指南。</p> <p><span>YOLOY8-0NNXRuntime-CPP:</span>与上述<span>ONNXRuntime</span>类似,但是是用<span>C++</span>编写的,包含了相应的<span>CMakeLists bt</span>,<span>inference.cpp.</span></p> <p><span>inference.h</span>和<span>main.cpp</span>文件,以及用于解释如何运行示例的<span>README.md</span>。</p> <p>每个示例都配有相应的文档,是当我们进行模型部署的时候在不同环境中部署和使用<span>YOLOv8</span>的示例<span>.</span></p> <p><strong><span>5.tests</span></strong></p> <p><span>conftest.py:</span>包含测试配置选项或共享的测试助手函数<span>test_cli.py:</span>用于测试命令行界面<span>(CLI)</span>的功能和行为。<span>test_cuda.py:</span>专门测试项目是否能正确使用<span>NVIDIA</span>的<span>CUDA</span>技术,确保<span>GPU</span>加速功能正常<span>test_engine.py:</span>测试底层推理引擎,如模型加载和数据处理等。<span>test integrations.py:</span>测试项目与其他服务或库的集成是否正常工作。<span>test_python.py:</span>用于测试项目的<span>Python AP|</span>接口是否按预期工作。</p> <p><strong><span>6.runs:</span></strong><strong><span>训练结果</span></strong></p> <p><strong><span>7.utralytics</span></strong></p> <p><span>datasets</span>文件夹<span>:</span>包含数据集的配置文件,如数据路径、类别信息等<span>(</span>就是我们训练<span>YOLO</span>模型的时候需要一个数据集,这里面就保存部分数据集的<span>yaml</span>文件,如果我们训练的时候没有指定数据集则会自动下载其中的数据集文件,但是很容易失败<span>!)</span>。<span>models</span>文件夹<span>:</span>存放模型配置文件,定义了模型结构和训练参数等,这个是我们改进或者就基础版本的一个<span>yam</span>文件配置的地方。</p> <p><span>nodelsx</span>件天中的每<span>yamlz</span>件代表了不同的<span>YOLOv8</span>模型配置,具体包括<span>:</span></p> <p><span>yolov8.yaml:</span>这是<span>YOLOv8</span>模型的标准配置文件,定义了模型的基础架构和参数。</p> <p><span>yolov8-cls.yam!: </span>配置文件调整了<span>YOLOv8</span>模型,专门用于图像分类任务。</p> <p><span>yolov8-ghost.yaml: </span>应用<span>Ghost</span>模块的<span>YOLOv8</span>变体,旨在提高计算效率<span>:yolov8-ghost-p2.yaml</span>和<span> yolov8-ghost-p6.yaml: </span>这些文件是针对特定大小输入的<span>Ghost</span>模型变体配置。<span>yolov8-p2.yaml</span>和<span> yolov8-p6.yam: </span>针对不同处理级别<span>(</span>例如不同的输入分辨率或模型深度<span>)</span>的<span>YOLOv8</span>模型配置</p> <p><span>yolov8-pose.yaml: </span>为姿态估计任务定制的<span>YOLOv8</span>模型配置。</p> <p><span>yolov8-pose-p6.yam!: </span>针对更大的输入分辨率或更复杂的模型架构姿态估计任务。</p> <p><span>yolov8-rtdetr.yaml: </span>可能表示实时检测和跟踪的<span>YOLOv8</span>模型变体。</p> <p><span>yolov8-seg.yaml</span>和<span> yolov8-seg-p6.yaml: </span>这些是为语义分割任务定制的<span>YOLOv8</span>模型配置</p> <p><span>trackers</span>文件夹<span>:</span>用于追踪算法的配置。</p> <p><span>init .py</span>文件<span>:</span>表明<span>`cfg</span>是<span>-</span>个<span>Python</span>包。</p> <p><span>default.yaml:</span>项目的默认配置文件,包含了被多个模块共享的通用配置项。</p> <p><strong><span>8.data</span></strong></p> <p><span>download weights.sh:</span>用来下载预训练权重的脚本。<span>get_coco.sh,get_coco128.sh,get imagenet.sh:</span>用于下载<span>COCO</span>数据集完整版、<span>128</span>张图片版以及<span>lmageNet</span>数据集的脚本</p> <p>在<span>data</span>文件夹中,包括<span>:</span></p> <p><span>annotator.py:</span>用于数据注释的工具。</p> <p><span>augment.py: </span>数据增强相关的函数或工具。<span>base.py,build.py,converter.py: </span>包含数据处理的基础类或函数、构建数据集的脚本以及数据格式转换工具</p> <p><span>dataset.py:</span>数据集加载和处理的相关功能。</p> <p><span>loaders.py: </span>定义加载数据的方法。</p> <p><span>utils.py:</span>各种数据处理相关的通用工具函数。</p> <p><strong><span>9.engine</span></strong></p> <p><span>exporter.py:</span>用于将训练好的模型导出到其他格式,例如<span>ONNX</span>或<span>TensorRT</span>。</p> <p><span>model.py: </span>包含模型定义,还包括模型初始化和加载的方法。</p> <p><span>predictor.py:</span>包含推理和预测的逻辑,如加载模型并对输入数据进行预测。</p> <p><span>results.py:</span>用于存储和处理模型输出的结果。</p> <p><span>trainer.py:</span>包含模型训练过程的逻辑。</p> <p><span>tuner.py: </span>用于模型超参数调优。</p> <p><span>validator.py: </span>包含模型验证的逻辑,如在验证集上评估模型性能。</p> <p><strong><span>10.models</span></strong></p> <p><span>classify:</span>这个目录可能包含用于图像分类的<span>YOLO</span>模型。</p> <p><span>detect: </span>包含用于物体检测的<span>YOLO</span>模型</p> <p><span>pose:</span>包含用于姿态估计任务的<span>YOLO</span>模型</p> <p><span>segment: </span>包含用于图像分割的<span>YOLO</span>模型</p> <p><strong><span>11.nn</span></strong></p> <p><span>modules</span>双件夹<span>:</span></p> <p><span>init .py: </span>表明此目录是<span>Python</span>包。</p> <p><span>block.py: </span>包含定义神经网络中的基础块,如残差块或瓶颈块。</p> <p><span>conv.py: </span>包含 卷积层<span>Q </span>相关的实现。</p> <p><span>head.py: </span>定义网络的头部,用于预测。</p> <p><span>transformer.py:</span>包含<span>Transformer</span>模型相关的实现</p> <p><span>utils.py: </span>提供构建神经网络时可能用到的辅助函数<span>,</span></p> <p><span>init .py:</span>同样标记这个目录为<span>Python</span>包</p> <p><span>autobackend.py: </span>用于自动选择最优的计算后端<span>,</span></p> <p><span>tasks,py,</span>定义了使用神经网络完<span>,</span>成的不同任务的流程,例如分类、检测或分割,所有的流程基本上都定义在这里,定义模型前向传播都在这里。</p> <p><strong><span>12.solutions</span></strong></p> <p><span>init_.py: </span>标识这是<span>-</span>个<span>Python</span>包。</p> <p><span>ai_gym.py:</span>与强化学习相关,例如在<span>OpenAlGym</span>环境中训练模型的代码<span>,heatmap.py:</span>用于生成和处理热图数据,这在物体检测和事件定位中很常见。<span>object counter.py: </span>用于物体计数的脚本,包含从图像中检测和计数实例的逻辑。</p> <p><strong><span>13.utils</span></strong></p> <p><span>callbacks.py:</span>包含在训练过程中被调用的回调函数。<span>autobatch.py:</span>用于实现批处理优化,以提高训练或推理的效率<span>:benchmarks.py: </span>包含性能基准测试相关的函数<span>.checks.py </span>用于项目中的各种检查,如参数验证或环境检查。<span>dist.py:</span>涉及分布式计算相关的工具</p> <p><span>downloads.py:</span>包含下载数据或模型等资源的脚本,<span>errors.py:</span>定义错误处理相关的类和函数<span>,</span></p> <p><span>fles.py: </span>包含文件操作相关的工具函数。</p> <p><span>instance.py: </span>包含实例化对象或模型的工具<span>,</span></p> <p><span>loss.py: </span>定义损失函数<span>,</span></p> <p><span>metrics.py: </span>包含评估模型性能的指标计算函数。</p> <p><span>ops.py: </span>包含自定义操作,如特殊的数学运算或数据转换<span>patches.py:</span>用于实现修改或补丁应用的工具<span>plotting.py: </span>包含数据可视化相关的绘图工具。</p> <p><span>tal.py:</span>一些损失函数的功能应用</p> <p><span>torch utils.py:</span>提供<span>PyTorch</span>相关的工具和辅助函数,包括<span>GFLOPs</span>的计算<span>triton.py: </span>可能与<span>NVIDlA Triton Inference Server</span>集成相关<span>tuner.py: </span>包含模型或算法调优相关的工具。</p> <p><strong><span>3.</span></strong><strong><span>模型模型</span></strong></p> <p>本研究模型主要在<span>YOLOv8s</span>主干,颈部和检测头部分进行改进,将<span>MobileNetV3</span>替换原始主干网络,<span>MobileNetV31.</span>结合了硬件感知的网络架构搜索<span>(NAS)</span>和<span>NetAdapt</span>算法,针对移动设备<span>CPU</span>进行优化<span>,</span>引入了新颖的架构设计,包括反转残差结构和线性瓶颈层。提出了高效的<span>Lite Reduced Atrous $patial Pyramid Pooling(LR-ASPP)</span>作为新的分割解码。</p> <p>在颈部部分引入<span>BiFF</span>双向金字塔网络,<span>BiFF</span>有高效的双向跨尺度连接<span>:BIFPN</span>通过在自顶向下和自底向上路径之间建立双向连接,允许不同尺度特征间的信息更有效地流动和融合。简化的网络结构<span>:BIFPN</span>通过删除只有一个输入边的节点、在同一层级的输入和输出节点间添加额外边,以及将每个双向路径视为<span>-</span>特征网络层并重复多次,来优化跨尺度连接。加权特征融合<span>:BIFPN</span>引入了可学习的权重来确定不同输入特征的重要性,从而提高了特征融合的效果。</p> <p>在检测头部分引入<span>ASFF</span>思想:自适应空间特征融合<span>:</span>提出了一种新的金字塔特征融合策略,能够空间过滤<span>,</span>中突信息,压制不同尺度特征间的不一致性。改善尺度不变性<span>:</span>通过<span>ASFF</span>策略,显著提升了特征的尺度不变性,有助于提高对象检测的准确性。低推理开销<span>:</span>在提升检测性能的同时,几乎不增加额外的推理开销。</p> <p><span> </span></p>
Pesticide risk during commercial apple pollination is greater for honeybees than other managed and wild bees
<p>Data and code relating to the manuscript “Pesticide risk during commercial apple pollination is greater for honeybees than other managed and wild bees”</p> <p>Files are organized as follows</p> <p><strong>input</strong> - contains the main data files.</p> <ul> <li> <p>all_pesticide2019.csv – contains the pesticide residue data for all samples</p> </li> <li> <p>LD50.csv – contains pesticide LD50s in PPB</p> </li> <li> <p>ld50_per_bee.csv – contains pesticide LD50s in ug / honeybee</p> </li> <li> <p>nesting_type.csv – lists nesting types of the different bee samples</p> </li> <li> <p>pesticide_type.csv – lists pesticides by their type</p> </li> <li> <p>short_name.csv – lists short bee names</p> <p><strong>folder “gis”</strong></p> <ul> <li>hive_distance_matrix.csv – distances from each orchard to other sites</li> <li>sitelocation.csv – contains coordinates of sites</li> </ul> </li> </ul> <p><strong>code</strong> - contains the scripts to analyze the input files</p> <ul> <li>pesticide_analysis.R – is the main anaylsis for the paper</li> <li>unadjusted_pesticide_analysis.R – is a copy of most of the code above but without LD50 weight adjustements</li> </ul> <p><strong>ld50 adjust</strong> - contains the inputs and code for the ld50 adjustments I ran</p> <ul> <li>ld50_adjust.R – is the code to calculate our the adjustements</li> <li>2020_BeeTox_database_acute_contact_publication_final_R1.csv – is the data file taken from pamminger publication</li> </ul>
Automated Picking Physical Load Exploration (APPLE) dataset
<p>This dataset provides physical data related to robotic apple picking. The data was collected in the autumn of 2023 on three separate orchard plots. We instrumented both the robot and the environment and collected synchronizable time series sensor data during robotic fruit picking. </p>
Combined effects of insecticide and IGP on native and invasive ladybeetles in apple orchard
<p>Raw data on the comparison of the combined effects of Rimon and IGP on two ladybeetle species</p>
Generation-dependent functional and numerical responses of Neoseiulus californicus (Phytoseiidae) long-term reared on thorn apple pollen
<p>Raw data of Functional and numerical response of N. califrnicus on T. urticae in different generaaations</p>
Supplementary Material | Dividing Apples and Pears: Towards a Taxonomy for Agile Transformation
<p><em>Agile Transformation (AT), the process of adopting agile methods and practices in organizational settings, has received grappling attention in research due to its extensive emergence in practice. Although the complexity of ATs is well known and its use cases widespread, research has not yet developed a comprehensive classification of AT. This lack limits the comparability of existing studies and our possibility to draw theoretical generalizations from their results. In this paper, we fill this gap by presenting a taxonomy for AT based on a systematic literature review. We abstracted the taxonomy in an analysis of 92 articles, including empirical and theoretical papers as well as experience reports. We contribute to the existing literature by providing a taxonomy that presents an analytical theory, offering a characterization of ATs which helps researchers and practitioners analyze ATs, identify how they differ, and provide insight into combinations of agile characteristics.</em></p>
Morphological and olfactory tree traits influence the susceptibility and suitability of the apple species Malus domestica and M. sylvestris to Anthonomus pomorum
<p><span>The florivorous apple blossom weevil, <em>Anthonomus pomorum</em> (Coleoptera: Curculionidae), is the most economically relevant insect pest of European apple orchards in early spring. Neither efficient monitoring nor ecologically sustainable management of this insect pest has yet been implemented. To identify heritable traits of apple trees that might influence host selection of <em>A. pomorum</em>, we compared susceptibility of apple tree species using infestation rates of the domesticated apple, <em>Malus domestica</em> (Rosaceae: Pyreae), and the European crab apple, <em>M. sylvestris</em>. We evaluated the suitability of the two apple species for <em>A. pomorum</em> by quantifying the mass of weevil offspring. Because volatile organic compounds (VOCs) emitted from flower buds of domesticated apple have previously been suggested to mediate female weevil preference via olfactory cues, we conducted bioassay experiments with blossom buds of both apple species to explore the olfactory preference of adult weevils and, furthermore, identified the headspace VOCs of blossom buds of both apple species through GC-MS analysis. The infestation analysis showed that<em> A. pomorum</em> infested the native European crab apple stronger than the domesticated apple, which originated from Central Asia. The European crab apple also appeared to be better suited for weevil larval development than the domesticated apple, as weevils emerging from <em>M. sylvestris</em> had a higher body mass than those emerging from <em>M. domestica</em>. These field observations were supported by olfactory bioassays, which showed that <em>A. pomorum</em> significantly preferred the odor of <em>M. sylvestris</em> buds compared to the odor of <em>M. domestica</em> buds. The analysis of headspace VOCs indicated differences in the blossom bud volatiles separating several <em>M. domestica</em> individuals from <em>M. sylvestris</em> individuals. This knowledge might be employed in further studies to repel <em>A. pomorum</em> from <em>M. domestica</em> blossom buds.</span></p> <p> </p>
Figure 2 in New Record Of The Invasive Channeled Apple Snail Pomacea Canaliculata (Lamarck, 1829) In Central Chile
Figure 2. Mature channeled apple snails and egg clutches found in Laguna del Muelle at Parque Bicentenario. A) Living mature, B) dorsal and ventral/apertural view of a shell found in the shore of the artificial lagoon, C) egg clutches attached to rocks and Alisma plantago- aquatica, and D) detail of an egg clutch. Scale bar = 1 cm.
Figure 1 in New Record Of The Invasive Channeled Apple Snail Pomacea Canaliculata (Lamarck, 1829) In Central Chile
Figure 1.Map showing previous and new record of Pomacea canaliculata in Chile. In detail, area surveyed at Bicentennial Park: The black square indicates Laguna del Muelle, where channeled apple snails were found and collected. Scale bar represents 1000 m.
Vineyard and Apple Orchard Suitability Maps for Mountainous Areas (Southern Pyrenees and Pre-Pyrenees)
<p>The manuscript related to this dataset can be consulted trought:</p> <p>The layers available in this dataset are in EPSG: WGS84.</p> <ul> <li><strong>Indicators</strong> <ul> <li><strong>BBL.tif </strong>- Hydrothermic index of Branas, Bernon, Levadoux (ºC*mm)</li> <li><strong>CDls.tif </strong>- Cold Days late spring (days)</li> <li><strong>FRea.tif </strong>- Frost Risk early autumn (days)</li> <li><strong>FRls_vineyard.tif </strong>- Frost Risk late spring vineyard (days)</li> <li><strong>FRls_apple.tif </strong>- Frost Risk late spring apple orchard (days)</li> <li><strong>GDD.tif</strong> - Growing Degree Days (ºC)</li> <li><strong>GSP.tif </strong>- Growing Season Precipitation (mm)</li> <li><strong>GST.tif </strong>- Growing Season Temperature (ºC)</li> <li><strong>Ha</strong><strong>.tif </strong>- Hail (days)</li> <li><strong>HI</strong><strong>.tif </strong>- Heliothermal Index of Huglin (ºC)</li> <li><strong>NCIr</strong><strong>.tif </strong>- Night Cool Index ripenning (ºC)</li> <li><strong>NHN.tif</strong> - Need Hydric Needs (mm/year)</li> <li><strong>SHDr.tif </strong>- Stressful Hot Days ripening (days)</li> <li><strong>WI.tif </strong>- Winkler Index (ºC)</li> <li><strong>CaCO3.tif </strong>- Calcium Carbonates (%)</li> <li><strong>CEC.tif </strong>- Cation Exchange Capacity (cmol/kg)</li> <li><strong>pH.tif </strong>- pH</li> <li><strong>SD.tif </strong>- Soil Depth (cm)</li> <li><strong>TAW.tif</strong> - Total Available Water (mm)</li> <li><strong>Te.tif</strong> - Texture</li> <li><strong>TOC.tif </strong>- Topsoil Organic Carbon (%)</li> <li><strong>As.tif </strong>- Aspect</li> <li><strong>GSR.tif </strong>- Growing season Solar Radiation (kWh/m2)</li> <li><strong>Sl.tif </strong>- Slope (%)</li> </ul> </li> </ul> <ul> <li><strong>Indicators_Suitability</strong> <ul> <li><strong>BBL_suitability.tif </strong>- Hydrothermic index of Branas, Bernon, Levadoux (ºC*mm)</li> <li><strong>CDls_suitability.tif </strong>- Cold Days late spring (days)</li> <li><strong>FRea_suitability.tif </strong>- Frost Risk early autumn (days)</li> <li><strong>FRls_vineyard_suitability.tif </strong>- Frost Risk late spring vineyard (days)</li> <li><strong>FRls_apple_suitability.tif </strong>- Frost Risk late spring apple orchard (days)</li> <li><strong>GDD_suitability.tif</strong> - Growing Degree Days (ºC)</li> <li><strong>GSP_suitability.tif </strong>- Growing Season Precipitation (mm)</li> <li><strong>GST_suitability.tif </strong>- Growing Season Temperature (ºC)</li> <li><strong>Ha_suitability</strong><strong>.tif </strong>- Hail (days)</li> <li><strong>HI_suitability</strong><strong>.tif </strong>- Heliothermal Index of Huglin (ºC)</li> <li><strong>NCIr_suitability</strong><strong>.tif </strong>- Night Cool Index ripenning (ºC)</li> <li><strong>NHN_suitability.tif</strong> - Need Hydric Needs (mm/year)</li> <li><strong>SHDr_suitability.tif </strong>- Stressful Hot Days ripening (days)</li> <li><strong>WI_suitabilitytif</strong> - Winkler Index (ºC)</li> <li><strong>CaCO3_suitability.tif </strong>- Calcium Carbonates (%)</li> <li><strong>CEC_suitability.tif </strong>- Cation Exchange Capacity (cmol/kg)</li> <li><strong>pH_suitability.tif </strong>- pH</li> <li><strong>SD_suitability.tif </strong>- Soil Depth (cm)</li> <li><strong>TAW_suitability.tif</strong> - Total Available Water (mm)</li> <li><strong>Te_suitability.tif</strong> - Texture</li> <li><strong>TOC_suitability.tif </strong>- Topsoil Organic Carbon (%)</li> <li><strong>As_suitability.tif </strong>- Aspect</li> <li><strong>GSR_suitability.tif </strong>- Growing season Solar Radiation (kWh/m2)</li> <li><strong>Sl_suitability.tif </strong>- Slope (%)</li> </ul> </li> </ul> <ul> <li><strong>Suitability</strong> <ul> <li><strong>Vineyard_suitability.tif </strong>- Vineyard Suitability map (Minumum Suitability 0 - 100 Maximum Suitability)</li> <li><strong>Apple_orchard_suitability.tif </strong>- Apple Orchard Suitability map (Minumum Suitability 0 - 100 Maximum Suitability) </li> </ul> </li> </ul>
Olfactory responses of Adalia bipunctata towards rosy apple aphid and its host plants
<p>dataset for the paper "<strong>Olfactory responses of <em>Adalia bipunctata</em> towards rosy apple aphid and its host plants"</strong></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.