A Microbiological Image Repository of Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa and Staphylococcus aureus Bacterial Colonies on Brain-Heart-Infusion Agar and Pseudomonas aeruginosa Bacterial Colonies on MacConkey Agar

This dataset consists of images of four types of bacterial strains. The images of the bacterial colonies were taken under two different shooting conditions, “controlled” and “uncontrolled” as described in “steps to reproduce” section. These bacterial strains are:
1.Pseudomonas aeruginosa (P. aeruginosa) on MacConkey agar: the number of images under controlled conditions is (376) and the number of images under uncontrolled conditions is (2078).
2.Staphylococcus aureus (S. aureus) on Brain-Heart-Infusion agar: the number of images under controlled conditions is (153) and the number of images under uncontrolled conditions is (3573).
3.Pseudomonas aeruginosa (P. aeruginosa) on Brain-Heart-Infusion agar: the number of images under controlled conditions is (89) and the number of images under uncontrolled conditions is (2735).
4.Klebsiella pneumoniae (K. pneumoniae) on Brain-Heart-Infusion agar: the number of images under controlled conditions is (158) and the number of images under uncontrolled conditions is (2579).
5.Escherichia coli (E. coli) on Brain-Heart-Infusion agar: the number of images under controlled conditions is (173) and the number of images under uncontrolled conditions is (2888).
IN THE REPOSITORY, YOU WILL FIND:
Group 3 consists of images taken from 24 plates of P. aeruginosa on MacConkey agar under controlled and uncontrolled conditions.
Group4 consists of images taken from:
25 plates of S. aureus, K. pneumoniae and E. coli on Brain-Heart-Infusion agar under controlled and uncontrolled conditions.
21 plates of P. aeruginosa on Brain-Heart-Infusion agar under controlled and uncontrolled conditions.
An excel sheet detailing the numbers of images in the folders.
NOTABLE FINDING:
Baseline CNNs trained on this data achieved high accuracy, indicating that phone images provide sufficient discriminative signal without expert inspection. Refer to references below.
HOW THIS DATASET CAN BE USED:
This dataset can be utilized in any research interested in recognizing different features of bacterial colonies. The dataset can also be used to train and evaluate deep learning models for colony classification, while also supporting studies on robustness, generalization, and practical deployment, to advance computer vision and AI applications in microbiology. By combining clinically important bacteria with controlled and unconstrained imaging, the dataset offers a realistic and accessible resource for researchers interested in developing AI methods that perform reliably in laboratory and non-laboratory environments.
IMPORTANT NOTE:
This dataset is an extension of a previous dataset which can be found at [https://doi.org/10.17632/kx6gz3wmcf.1]. The expansion includes additional bacterial strains and culture media.
IF YOU USE THIS DATASET, PLEASE REFERENCE THE FOLLOWING:
DOI: https://doi.org/10.1109/ACCESS.2022.3221958
DOI: https://doi.org/10.1109/ACCESS.2025.3625648
DOI: https://doi.org/10.17632/kx6gz3wmcf.1
DOI: https://doi.org/10.17632/v54x8jdx5x.1