A spectral learning path for simultaneous multi-parameter detection of water quality

A novel spectroscopy-deep learning approach for aqueous multi-heavy metal detection

Addressing heavy metal contamination in water bodies is a critical concern for environmental scientists. Traditional detection methods are often complex and costly. Recent advancements in artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), have shown significant potential in analytical chemistry. However, these AI models require extensive spectral data, which traditional methods struggle to provide quickly. To overcome this challenge, we developed a new digital spectral imaging system and rapidly collected 3000 digital spectra from mixed heavy metal samples. We then created an end-to-end regression model for predicting heavy metal concentrations in mixed water samples using deep convolutional neural networks (ResNet-50, Inception V1, and SqueezeNet V1.1). The results indicated that the trained ResNet-50 model can effectively detect arsenic, chromium, and copper simultaneously, with a linear fitting coefficient exceeding 0.99 between true and predicted values. This study offers an efficient approach for rapid heavy metal detection in complex water environments and serves as a reference for developing intelligent analytical techniques.

A novel method for multi-pollutant monitoring in water supply systems using chemical machine vision

Drinking water is vital for human health and life, but detecting multiple contaminants in it is challenging. Traditional testing methods are both time-consuming and labor-intensive, lacking the ability to capture abrupt changes in water quality over brief intervals. This paper proposes a direct analysis and rapid detection method of three indicators of arsenic, cadmium, and selenium in complex drinking water systems by combining a novel long-path spectral imager with machine learning models. Our technique can obtain multiple parameters in about 1 s. The experiment involved setting up samples from various drinking water backgrounds and mixed groups, totaling 9360 injections. A raw visible light source ranging from 380 to 780 nm was utilized, uniformly dispersing light into the sample cell through a filter. The residual beam was captured by a high-definition camera, forming a distinctive spectrum. Three deep learning models-ResNet-50, SqueezeNet V1.1, and GoogLeNet Inception V1-were employed. Datasets were divided into training, validation, and test sets in a 6:2:2 ratio, and prediction performance across different datasets was assessed using the coefficient of determination and root mean square error. The experimental results show that a well-trained machine learning model can extract a lot of feature image information and quickly predict multi-dimensional drinking water indicators with almost no preprocessing. The model's prediction performance is stable under different background drinking water systems. The method is accurate, efficient, and real-time and can be widely used in actual water supply systems. This study can improve the efficiency of water quality monitoring and treatment in water supply systems, and the method's potential for environmental monitoring, food safety, industrial testing, and other fields can be further explored in the future.