A machine vision dataset for automated quality inspection and grading of sweetpotatoes

Dataset Structure SP_MVDataset.7z ├─Subset A │ ├─ Sweetpotato Sampling Datasheet.docx │ │ │ ├─ Images │ │ Batch_01_Frame_001.png │ │ Batch_01_Frame_002.png │ │ ... │ │ Batch_19_Frame_012.png │ │ │ ├─ Labels │ │ Batch_01_Frame_001.json │ │ Batch_01_Frame_002.json │ │ ... │ │ Batch_19_Frame_012.json │ └─ Videos │ Batch_01.avi │ Batch_02.avi │ ... │ Batch_20.avi │ └─Subset B ├─ Images │ Batch_01_Frame_001.png │ Batch_01_Frame_002.png │ ... │ Batch_20_Frame_039.png │ ├─ Labels │ Batch_01_Frame_001.json │ Batch_01_Frame_002.json │ ... │ Batch_20_Frame_039.json │ └─ Videos Batch_01.avi Batch_02.avi ... Batch_20.avi Dataset Organization When extracting the data from the archive, the dataset is organized into two main components: Subset A: Contains data from 123 commercial sweetpotatoes (Grocery Store source) imaged under ambient indoor lighting. Subset B: Contains data from 267 fresh-harvested sweetpotatoes (Research Station source) imaged in an enclosed LED chamber. Additionally, a supplementary file "Sweetpotato_Sampling_Datasheet.docx" is included, providing population statistics (weight, length, width) and surface conditions for the samples in Subset A. For both subset, the raw video recordings were provided for further research. Directory Hierarchy Within the subset directories, the data is further organized into subfolders for images and annotations: `Subset_A/Images`: Contains 232 RGB frames (Resolution: 1920×1080 pixels). `Subset_A/Labels`: Contains 232 corresponding JSON annotation files. `Subset_A/Videos`: Contains 19 corresponding raw video files. `Subset_B/Images`: Contains 1168 RGB frames (Resolution: 1280×720 pixels). `Subset_B/Labels`: Contains 1168 corresponding JSON annotation files. `Subset_B/Videos`: Contains 20 corresponding raw video files Each image is a standard RGB .png file. The samples were rotated on a roller conveyor during acquisition to capture full-surface views. Dataset Summary - Total samples: 390 (123 in Subset A, 267 in Subset B) - Total images: 1,400 - Total annotated instances: 3,700 - Total videos: 39 - Storage space required: Approximately 6.54 GB (uncompressed) File Naming Convention The file naming convention is consistent across images and annotation files to ensure traceability to the original video batches. Each file name includes two key elements: 1. Batch ID: Represents the specific group or video sequence (e.g., "Batch_1"). 2. Frame Sequence: Represents the sequential order of the frame extracted (e.g., "Frame_001"). Examples - Batch_01_Frame_001.png: The 1st frame extracted from Batch 1 video. - Batch_01_Frame_001.json: The corresponding annotation file for the image above. Note: Files in Subset A and Subset B share this naming convention but are stored in separate parent directories to distinguish the domain/source. Annotation Structure The annotation files (`.json`) follow the standard LabelMe format. They are fully compatible with common computer vision tools (e.g., LabelMe, AnyLabeling). Each file contains: 1. Shapes: - label: The visual quality category (Grade 1, Grade 2, or Grade 3). - points: A list of [x, y] coordinates defining the polygon mask around the sweetpotato instance. - shape_type: "polygon". 2. Image Path: References the corresponding .png image file. Class Definitions Sweetpotato instances are labeled into three visual categories based on the visible surface defects in the specific frame: - Grade 1 (Normal): High-quality instances. Defect-free or negligible surface imperfections visible. - Grade 2 (Moderate Defects): Instances with visible surface defects that affect appearance (e.g., minor scuffs, skinning). - Grade 3 (Severe Defects): Instances with significant damage or decay visible (e.g., deep cuts, rot, severe mechanical damage). Usage Notes - Data Splitting: Users should split the dataset (Training/Testing) based on the Batch ID (Video level), not by individual frames. Randomly splitting frames will result in data leakage as adjacent frames capture the same samples. - ID Mapping: The dataset is designed for instance-level vision tasks (segmentation/detection). Instance-to-Sample ID mapping (tracking specific physical roots across frames) is not explicitly provided in the metadata. Citations If you use the dataset in published research, please consider citing related journal articles or the dataset. Xu, J., Lu, Y., & Deng, B. (2024). Design, prototyping, and evaluation of a new machine vision-based automated sweetpotato grading and sorting system, Journal of the ASABE 67 (2024) 1369–1380. https://doi.org/10.13031/ja.16051 Xu, J. & Lu, Y. (2024). Prototyping and evaluation of a novel machine vision system for real-time, automated quality grading of sweetpotatoes, Computers and Electronics in Agriculture 219 (2024) 108826. https://doi.org/10.1016/j.compag.2024.108826 Zhang, J., Lu, Y., & Xu, J. (2025). A machine vision dataset for automated quality inspection and grading of sweetpotatoes [Data set]. Zenodo. https://doi.org/10.5281/zenodo.18100484 Hopefully, you find the dataset useful.## 数据集结构 SP_MVDataset.7z ├─ 子集A（Subset A） │ ├─ 甘薯采样数据表（Sweetpotato Sampling Datasheet）.docx │ │ │ ├─ 图像（Images） │ │ Batch_01_Frame_001.png │ │ Batch_01_Frame_002.png │ │ ... │ │ Batch_19_Frame_012.png │ │ │ ├─ 标注（Labels） │ │ Batch_01_Frame_001.json │ │ Batch_01_Frame_002.json │ │ ... │ │ Batch_19_Frame_012.json │ └─ 视频（Videos） │ Batch_01.avi │ Batch_02.avi │ ... │ Batch_20.avi │ └─ 子集B（Subset B） ├─ 图像（Images） │ Batch_01_Frame_001.png │ Batch_01_Frame_002.png │ ... │ Batch_20_Frame_039.png │ ├─ 标注（Labels） │ Batch_01_Frame_001.json │ Batch_01_Frame_002.json │ ... │ Batch_20_Frame_039.json │ └─ 视频（Videos） Batch_01.avi Batch_02.avi ... Batch_20.avi ## 数据集组织方式 解压该归档文件后，数据集分为两个核心子集： - 子集A（Subset A）：包含123份来自杂货店的商业甘薯（sweetpotato）样本，在室内环境光下完成成像。 - 子集B（Subset B）：包含267份来自科研站的新鲜采收甘薯（sweetpotato）样本，在封闭LED舱内完成成像。 此外，归档中包含补充文件「甘薯采样数据表（Sweetpotato Sampling Datasheet）.docx」，提供了子集A样本的群体统计信息（重量、长度、宽度）及表面状态数据。两个子集均提供原始视频录制文件，以供后续研究使用。 ## 目录层级结构 在各子集目录下，数据进一步按子文件夹分类存储： - 子集A/图像（Subset_A/Images）：包含232帧RGB图像，分辨率为1920×1080像素。 - 子集A/标注（Subset_A/Labels）：包含232份对应的JSON标注文件。 - 子集A/视频（Subset_A/Videos）：包含19份对应的原始视频文件。 - 子集B/图像（Subset_B/Images）：包含1168帧RGB图像，分辨率为1280×720像素。 - 子集B/标注（Subset_B/Labels）：包含1168份对应的JSON标注文件。 - 子集B/视频（Subset_B/Videos）：包含20份对应的原始视频文件。 所有图像均为标准RGB格式的PNG文件。采集过程中，样本在滚筒传送带上旋转，以获取完整的表面视角。 ## 数据集概览 - 总样本数：390份（子集A含123份，子集B含267份） - 总图像数：1400帧 - 总标注实例数：3700个 - 总视频数：39份 - 所需存储空间：未压缩状态下约6.54 GB ## 文件命名规范 图像与标注文件采用统一命名规则，以确保可追溯至原始视频批次。每个文件名包含两个核心要素： 1. 批次ID（Batch ID）：代表特定分组或视频序列（例如"Batch_1"）。 2. 帧序号（Frame Sequence）：代表提取帧的顺序编号（例如"Frame_001"）。 命名示例： - "Batch_01_Frame_001.png"：从批次1视频中提取的第1帧图像。 - "Batch_01_Frame_001.json"：上述图像对应的标注文件。 注意：子集A与子集B均遵循此命名规范，但存储于不同的父目录中，以区分数据来源与应用域。 ## 标注文件结构 标注文件（.json）遵循标准LabelMe格式，可与常见计算机视觉工具（例如LabelMe、AnyLabeling）兼容。每份标注文件包含以下内容： 1. 形状信息（Shapes）： - 标签（label）：视觉质量等级（1级、2级或3级）。 - 顶点坐标（points）：定义甘薯实例周围多边形掩码的[x, y]坐标列表。 - 形状类型（shape_type）："polygon"（多边形）。 2. 图像路径：指向对应的.PNG图像文件。 ## 类别定义 根据单帧图像中可见的表面缺陷，甘薯实例被划分为三个视觉质量类别： - 1级（正常）：高品质实例，表面无缺陷或仅存在可忽略的细微瑕疵。 - 2级（轻度缺陷）：存在影响外观的可见表面缺陷的实例（例如轻微划痕、表皮剥落）。 - 3级（严重缺陷）：存在明显损伤或腐烂的实例（例如深度切口、腐烂、严重机械损伤）。 ## 使用注意事项 - 数据划分：用户应基于批次ID（视频级别）划分数据集（训练集/测试集），而非按单帧图像划分。若随机划分单帧，会导致数据泄露，因为相邻帧捕获的是同一甘薯样本。 - ID映射：本数据集专为实例级视觉任务（分割/检测）设计，元数据中未提供实例-样本ID的映射关系（即跨帧追踪特定物理薯块的信息）。 ## 引用说明 若您在已发表的研究中使用本数据集，请引用相关期刊文章或数据集本身： 1. Xu, J., Lu, Y., & Deng, B. (2024). Design, prototyping, and evaluation of a new machine vision-based automated sweetpotato grading and sorting system. *Journal of the ASABE*, 67, 1369–1380. https://doi.org/10.13031/ja.16051 2. Xu, J. & Lu, Y. (2024). Prototyping and evaluation of a novel machine vision system for real-time, automated quality grading of sweetpotatoes. *Computers and Electronics in Agriculture*, 219, 108826. https://doi.org/10.1016/j.compag.2024.108826 3. Zhang, J., Lu, Y., & Xu, J. (2025). A machine vision dataset for automated quality inspection and grading of sweetpotatoes [Data set]. Zenodo. https://doi.org/10.5281/zenodo.18100484 希望本数据集对您的研究有所助益。