BacteriaDetect

This project is the machine vision algorithm implementation of bacterial detection, which aims to detect Bacteria circles' location from biochip and output its CIE values in a rectangle area.

Article

Point-of-Care Pathogen Testing Using Photonic Crystals and Machine Vision for Diagnosis of Urinary Tract Infections https://pubs.acs.org/doi/abs/10.1021/acs.nanolett.0c04942

Requirement

python = 3.7.8
numpy = 1.19.1
opencv = 4.4.0

Creating a new conda environment is recommended. Run this script like: conda create -n test python=3.7 opencv numpy -c conda-forge

List of Arguments

Argument	Required/Not	Details
-h, --help	optional	show help message and exit.
-i, --in_dir	required	input directory (relative dir.) of images.
-o, --out_dir	optional	output dir of detected images.
-d, --min_circle_distance	optional	Minimum distance of adjacent circles(pixels). Default: 60
-e, --edge_detect_thres	optional	Contrast threshold between circle edge and background. Default: 26
-r, --roundness_thres	optional	Roundness threshold of circles. Default: 31
--min_circleRadius	optional	Minimum of circle radius. Default: 20
--max_circleRadius	optional	Maximum of circle radius. Default: 90

Usage

1. Put test images in the testdata/images.tif in the form of a folder.
2. Open anaconda/cmd prompt, activate the installed environment, and change the working directory to the current project.
3. When not specifying the output path, a folder named by runtime will be generated in the current directory by default.
4. The above five default detection parameters can be modified appropriately when the detection is incomplete.
5. Run python script like: python identify.py -i testdata/

Testdata and Results

Input image

Output results

Citation

Please consider citing the following article if you used this project in your research.

@article{liu2021point,
	title={Point-of-Care Pathogen Testing Using Photonic Crystals and Machine Vision for Diagnosis of Urinary Tract Infections},
	author={Liu, Haoran and Li, Zhihao and Shen, Ruichen and Li, Zhiheng and Yang, Yanbing and Yuan, Quan},
	journal={Nano Letters},
	volume={21},
	number={7},
	pages={2854--2860},
	year={2021},
	publisher={ACS Publications}
}