Researchers at NIT Rourkela have developed an AI-powered system that can detect and quantify adulteration in spices and other food products within seconds, offering a faster and more scalable alternative to conventional laboratory testing.

The system pairs Fourier Transform Infrared (FTIR) spectroscopy with machine learning algorithms to analyse food samples in real time. Unlike traditional methods such as chromatography, which are accurate but resource-heavy, the new system is non-destructive, cost-effective, and compatible with industrial quality control workflows.

In trials, the system achieved around 92% accuracy in detecting adulteration in coriander powder, including the presence of sawdust as an adulterant. Crucially, it can also quantify contamination levels, not merely confirm their presence.

“Our invention focuses on addressing a long-standing challenge in the food industry, which is the absence of a fast and reliable spice adulteration detection system,” said Prof. Sushil Kumar Singh, one of the lead researchers. “This innovation will not only ensure food safety and regulatory compliance but also strengthen consumer trust across the supply chain.”

The research, published in the journal Food Chemistry, was conducted by Prof. Singh, the late Prof. Poonam Singha, and M.Tech scholar Rishabh Goyal from the Department of Food Process Engineering. The team has also secured a patent (No. 581403) for the technology.