Related Experiment Video
Updated: Jun 19, 2026

10:21
Multicolor Fluorescence Detection for Droplet Microfluidics Using Optical Fibers
Published on: May 5, 2016
10.5K
Research on Water Quality Chemical Oxygen Demand Detection Using Laser-Induced Fluorescence Image Processing
Ying Guo1, Zhaoshuo Tian1, Zongjie Bi1
1Institute of Marine Optoelectronic Equipment, Harbin Institute of Technology at Weihai, Weihai 264209, China.
Sensors (Basel, Switzerland)
|March 17, 2025
Summary
A new Laser-Induced Fluorescence (LIF) image processing method accurately detects low-concentration Chemical Oxygen Demand (COD) in water. This non-contact system offers high sensitivity for real-time water quality monitoring.
Area of Science:
- Environmental Science
- Analytical Chemistry
- Optical Engineering
Background:
- Chemical Oxygen Demand (COD) is a key indicator of organic water pollution.
- Accurate measurement of low-concentration COD is essential for effective water quality management.
- Existing methods may have limitations in sensitivity or real-time application.
Purpose of the Study:
- To develop and validate an innovative Laser-Induced Fluorescence (LIF) image processing technique for detecting low-concentration COD in aqueous environments.
- To design and build a miniaturized, non-contact COD detection system based on LIF image processing.
- To assess the system's performance in terms of sensitivity, stability, and accuracy.
Main Methods:
- Utilized ultraviolet laser excitation to induce fluorescence in organic compounds within water samples.
- Employed a CMOS image sensor to capture fluorescence image data.
- Applied image processing techniques, including color channel isolation and RGB characteristic analysis, to derive COD values.
- Developed a predictive model using Partial Least Squares Regression (PLSR) based on image features.
Main Results:
- Successfully developed an LIF image processing system for COD detection, comprising a CMOS sensor, STM32 microprocessor, laser, and display.
- Acquired fluorescence images from mixed solutions of sodium humate and glucose at varying concentrations.
- Achieved a detection relative error of less than 10% for COD concentrations within the 0-12 mg/L range.
- Demonstrated the system's capability to correlate RGB color characteristics with COD concentrations.
Conclusions:
- The developed LIF image processing system offers a highly sensitive, stable, and non-contact method for measuring low-concentration COD.
- The system is suitable for rapid, real-time online water quality monitoring.
- This innovative approach provides a promising alternative for environmental water pollution assessment.
Keywords:
RGB color featureschemical oxygen demandimage processinglaser-induced fluorescencelow concentration
