Poultry Birds Poo Imagery Dataset for Health Status Prediction: A Case of South-West Nigeria
Description
Imagery dataset of chicken poo of both healthy and non-healthy collected from poultry farms rearing two species of chickens namely: broilers and layers in Nigeria. The dataset has a total of 14,618 labeled images with 100 × 100 pixels in jpeg format and a label indicating the name of the image from the image number. In the repository, data were uploaded into 2 separate folders; each folder of either healthy or non-healthy contain images taken from different time to the day; morning, afternoon, night and others.
Files
Steps to reproduce
Mobile device with 13 megapixels camera and connected to Edge Impulse Platform was used to collect image data. Data collection was done by both researchers and poultry farmers, while data quality check was done by Animal Scientist. Data collection was done in three months, from February to April 2023, and involved poultry farms on Bowen University campus and Iwo regions.