maxKudi/Cell-Dataset

The Cell dataset contains a total of 2500 images of living cells and dead cells with annotations.

Cell-Dataset

The Living and Dead cell (LDC) dataset contains 2500 images along with their annotation files, the original dataset is not split into Training, Testing, and Validation sets. I would recommend using roboflow website to augment the dataset. Among the 2500 images, Annotated Living Cells are 7003 while Dead Cells are 1516.

Performance

The Dataset still have some noticeable error so i would recommend picking the best images among the 2500 before performing the augmentation

Data Description

Image

Each image is resized to 225 x 219 resolution.

N.B. Rectangular bounding boxes are converted to circular bounding boxes for representation.

Annotation Format

<annotation>
	<folder>Label</folder>
	<filename>b (568).png</filename>
	<path>C:\Users\Kudi Okerulu\Desktop\Label\b (568).png</path>
	<source>
		<database>Unknown</database>
	</source>
	<size>
		<width>225</width>
		<height>219</height>
		<depth>1</depth>
	</size>
	<segmented>0</segmented>
	<object>
		<name>Living Cell</name>
		<pose>Unspecified</pose>
		<truncated>0</truncated>
		<difficult>0</difficult>
		<bndbox>
			<xmin>157</xmin>
			<ymin>83</ymin>
			<xmax>178</xmax>
			<ymax>105</ymax>
		</bndbox>
	</object>
</annotation>