U<sup>2</sup>-Net and ResNet50-Based Automatic Pipeline for Bacterial Colony Counting
In this paper, an automatic colony counting system based on an improved image preprocessing algorithm and convolutional neural network (CNN)-assisted automatic counting method was developed. Firstly, we assembled an LED backlighting illumination platform as an image capturing system to obtain photog...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Microorganisms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-2607/12/1/201 |
| _version_ | 1797339588084105216 |
|---|---|
| author | Libo Cao Liping Zeng Yaoxuan Wang Jiayi Cao Ziyu Han Yang Chen Yuxi Wang Guowei Zhong Shanlei Qiao |
| author_facet | Libo Cao Liping Zeng Yaoxuan Wang Jiayi Cao Ziyu Han Yang Chen Yuxi Wang Guowei Zhong Shanlei Qiao |
| author_sort | Libo Cao |
| collection | DOAJ |
| description | In this paper, an automatic colony counting system based on an improved image preprocessing algorithm and convolutional neural network (CNN)-assisted automatic counting method was developed. Firstly, we assembled an LED backlighting illumination platform as an image capturing system to obtain photographs of laboratory cultures. Consequently, a dataset was introduced consisting of 390 photos of agar plate cultures, which included 8 microorganisms. Secondly, we implemented a new algorithm for image preprocessing based on light intensity correction, which facilitated clearer differentiation between colony and media areas. Thirdly, a U<sup>2</sup>-Net was used to predict the probability distribution of the edge of the Petri dish in images to locate region of interest (ROI), and then threshold segmentation was applied to separate it. This U<sup>2</sup>-Net achieved an F1 score of 99.5% and a mean absolute error (MAE) of 0.0033 on the validation set. Then, another U<sup>2</sup>-Net was used to separate the colony region within the ROI. This U<sup>2</sup>-Net achieved an F1 score of 96.5% and an MAE of 0.005 on the validation set. After that, the colony area was segmented into multiple components containing single or adhesive colonies. Finally, the colony components (CC) were innovatively rotated and the image crops were resized as the input (with 14,921 image crops in the training set and 4281 image crops in the validation set) for the ResNet50 network to automatically count the number of colonies. Our method achieved an overall recovery of 97.82% for colony counting and exhibited excellent performance in adhesion classification. To the best of our knowledge, the proposed “light intensity correction-based image preprocessing→U<sup>2</sup>-Net segmentation for Petri dish edge→U<sup>2</sup>-Net segmentation for colony region→ResNet50-based counting” scheme represents a new attempt and demonstrates a high degree of automation and accuracy in recognizing and counting single-colony and multi-colony targets. |
| first_indexed | 2024-03-08T09:50:24Z |
| format | Article |
| id | doaj.art-fc88746c81364b958f3d6a092a91e8d9 |
| institution | Directory Open Access Journal |
| issn | 2076-2607 |
| language | English |
| last_indexed | 2024-03-08T09:50:24Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Microorganisms |
| spelling | doaj.art-fc88746c81364b958f3d6a092a91e8d92024-01-29T14:06:58ZengMDPI AGMicroorganisms2076-26072024-01-0112120110.3390/microorganisms12010201U<sup>2</sup>-Net and ResNet50-Based Automatic Pipeline for Bacterial Colony CountingLibo Cao0Liping Zeng1Yaoxuan Wang2Jiayi Cao3Ziyu Han4Yang Chen5Yuxi Wang6Guowei Zhong7Shanlei Qiao8Center for Global Health, Nanjing Medical University, Nanjing 211166, ChinaDepartment of Pathogen Biology, School of Basic Medical Sciences, Nanjing Medical University, Nanjing 211166, ChinaCenter for Global Health, Nanjing Medical University, Nanjing 211166, ChinaCenter for Global Health, Nanjing Medical University, Nanjing 211166, ChinaCenter for Global Health, Nanjing Medical University, Nanjing 211166, ChinaCenter for Global Health, Nanjing Medical University, Nanjing 211166, ChinaCenter for Global Health, Nanjing Medical University, Nanjing 211166, ChinaCenter for Global Health, Nanjing Medical University, Nanjing 211166, ChinaCenter for Global Health, Nanjing Medical University, Nanjing 211166, ChinaIn this paper, an automatic colony counting system based on an improved image preprocessing algorithm and convolutional neural network (CNN)-assisted automatic counting method was developed. Firstly, we assembled an LED backlighting illumination platform as an image capturing system to obtain photographs of laboratory cultures. Consequently, a dataset was introduced consisting of 390 photos of agar plate cultures, which included 8 microorganisms. Secondly, we implemented a new algorithm for image preprocessing based on light intensity correction, which facilitated clearer differentiation between colony and media areas. Thirdly, a U<sup>2</sup>-Net was used to predict the probability distribution of the edge of the Petri dish in images to locate region of interest (ROI), and then threshold segmentation was applied to separate it. This U<sup>2</sup>-Net achieved an F1 score of 99.5% and a mean absolute error (MAE) of 0.0033 on the validation set. Then, another U<sup>2</sup>-Net was used to separate the colony region within the ROI. This U<sup>2</sup>-Net achieved an F1 score of 96.5% and an MAE of 0.005 on the validation set. After that, the colony area was segmented into multiple components containing single or adhesive colonies. Finally, the colony components (CC) were innovatively rotated and the image crops were resized as the input (with 14,921 image crops in the training set and 4281 image crops in the validation set) for the ResNet50 network to automatically count the number of colonies. Our method achieved an overall recovery of 97.82% for colony counting and exhibited excellent performance in adhesion classification. To the best of our knowledge, the proposed “light intensity correction-based image preprocessing→U<sup>2</sup>-Net segmentation for Petri dish edge→U<sup>2</sup>-Net segmentation for colony region→ResNet50-based counting” scheme represents a new attempt and demonstrates a high degree of automation and accuracy in recognizing and counting single-colony and multi-colony targets.https://www.mdpi.com/2076-2607/12/1/201bacterial colony countinglight intensity correctionconvolutional neural networksU<sup>2</sup>-NetResNet50image segmentation |
| spellingShingle | Libo Cao Liping Zeng Yaoxuan Wang Jiayi Cao Ziyu Han Yang Chen Yuxi Wang Guowei Zhong Shanlei Qiao U<sup>2</sup>-Net and ResNet50-Based Automatic Pipeline for Bacterial Colony Counting Microorganisms bacterial colony counting light intensity correction convolutional neural networks U<sup>2</sup>-Net ResNet50 image segmentation |
| title | U<sup>2</sup>-Net and ResNet50-Based Automatic Pipeline for Bacterial Colony Counting |
| title_full | U<sup>2</sup>-Net and ResNet50-Based Automatic Pipeline for Bacterial Colony Counting |
| title_fullStr | U<sup>2</sup>-Net and ResNet50-Based Automatic Pipeline for Bacterial Colony Counting |
| title_full_unstemmed | U<sup>2</sup>-Net and ResNet50-Based Automatic Pipeline for Bacterial Colony Counting |
| title_short | U<sup>2</sup>-Net and ResNet50-Based Automatic Pipeline for Bacterial Colony Counting |
| title_sort | u sup 2 sup net and resnet50 based automatic pipeline for bacterial colony counting |
| topic | bacterial colony counting light intensity correction convolutional neural networks U<sup>2</sup>-Net ResNet50 image segmentation |
| url | https://www.mdpi.com/2076-2607/12/1/201 |
| work_keys_str_mv | AT libocao usup2supnetandresnet50basedautomaticpipelineforbacterialcolonycounting AT lipingzeng usup2supnetandresnet50basedautomaticpipelineforbacterialcolonycounting AT yaoxuanwang usup2supnetandresnet50basedautomaticpipelineforbacterialcolonycounting AT jiayicao usup2supnetandresnet50basedautomaticpipelineforbacterialcolonycounting AT ziyuhan usup2supnetandresnet50basedautomaticpipelineforbacterialcolonycounting AT yangchen usup2supnetandresnet50basedautomaticpipelineforbacterialcolonycounting AT yuxiwang usup2supnetandresnet50basedautomaticpipelineforbacterialcolonycounting AT guoweizhong usup2supnetandresnet50basedautomaticpipelineforbacterialcolonycounting AT shanleiqiao usup2supnetandresnet50basedautomaticpipelineforbacterialcolonycounting |