BCCD

该数据集名为BCCD，包含364张图像，每张图像的尺寸为640×480×3，其中包含了不同比例的血小板、白细胞和红细胞。为了实验需要，所有图像都被调整为416×416×3的大小。数据集按照7:2:1的比例被划分为训练集、验证集和测试集，所有细胞均以COCO格式进行标注。此外，为了增强数据，还采用了水平翻转、垂直翻转、随机裁剪、亮度调整和曝光调整等数据增强技术。该数据集的规模为364张图像，任务是对血液细胞进行检测。This dataset, named BCCD, consists of 364 images with an original resolution of 640×480×3, which contain platelets, leukocytes, and erythrocytes in varying proportions. For experimental purposes, all images were resized to 416×416×3. The dataset is split into training, validation, and test sets at a ratio of 7:2:1, with all cells annotated in COCO format. Additionally, data augmentation techniques including horizontal flipping, vertical flipping, random cropping, brightness adjustment, and exposure adjustment were employed to enhance the dataset. With 364 images in total, the core task of this dataset is blood cell detection.

搜集汇总

背景与挑战

背景概述

BCCD数据集是一个小规模的血液细胞检测数据集，包含364张640x480像素的JPEG图像，标注为VOC格式的XML文件，涵盖红细胞、白细胞和血小板三种细胞类型。数据集还提供了数据准备脚本和mxnet预处理脚本，适用于机器学习和目标检测任务。