CongNaMul

CongNaMul是一个专为大豆芽图像分析设计的综合数据集，旨在支持图像分类、语义分割、分解和长度及重量测量等任务。分类任务提供四个类别以评估大豆芽的质量：正常、破损、斑点和破损带斑点，用于开发AI辅助的自动质量检测技术。语义分割部分包含从单个芽到多个芽的图像，以及人工标注的掩码图像，标签有4个不同类别：背景、头部、身体、尾部。数据集还提供图像和掩码，用于图像分解任务，包括两个单独的芽图像及其组合形式。最后，提供5个芽的物理特征（头部长度、身体长度、身体厚度、尾部长度、重量）用于基于图像的测量任务。该数据集预计将成为大豆芽图像高级分析研究的宝贵资源，并希望帮助其他工业领域的研究人员评估其模型。

CongNaMul is a comprehensive dataset specifically designed for soybean sprout image analysis, aiming to support tasks including image classification, semantic segmentation, image decomposition, as well as length and weight measurement tasks. For the classification task, four categories are provided to evaluate the quality of soybean sprouts: normal, damaged, spotted, and damaged with spots, which is used to develop AI-assisted automatic quality detection technologies. The semantic segmentation section includes images ranging from single sprouts to multiple sprouts, alongside manually annotated mask images, with four distinct label categories: background, head, body, and tail. The dataset also provides images and masks for the image decomposition task, covering two individual sprout images and their combined form. Finally, physical features of five soybean sprouts (head length, body length, body thickness, tail length, and weight) are provided for image-based measurement tasks. This dataset is expected to serve as a valuable resource for advanced analysis research on soybean sprout images, and is intended to help researchers in other industrial fields evaluate their models.