SpiceSafeNet: An Explainable AI System for Robust Detection of Red Chilli Powder Adulteration

Rifat F. M. Solkar, Fatima T. A. Tandel and Harshada U. Salvi

International Journal of Scientific Research in Science and Technology, 2026, vol. 13, issue 3, 962-971

Abstract: The adulteration of red chilli powder, a common kitchen ingredient in every household of India, by means of substituting chilli with brick powder is a matter of great concern for health-conscious buyers, as there are limited methods for a layman to test whether the chilli powder he buys consists of only the given spices, or of inferior quality adulterants as well. Brick powder is commonly added in to chilli powder, as it can be easily made to resemble chilli powder both in terms of colour and texture. These mixtures, when consumed for a considerable length of time can result in the buyer being suffering from long-term damages and digestive system related diseases. While methods like chromatography and spectrometry give authentic readings, they only serve in laboratories due to being expensive, and requiring technically skilled personnel. This paper describes a software system called SpiceSafeNet, which detects presence of brick powder adulterant in pictures of red chilli powder that can be clicked using any normal camera. The system, utilizing a ResNet-18 deep learning model, trained on a dataset of 493 images of both adulterated and un-adulterated samples using transfer learning. An image, upon being uploaded, goes through a test of being 'red chilli powder' by checking its ratio of red colored pixels, before being fed to the model which produces a predicted output, with the confidence with which the prediction has been made, ranging from 0 to 1.0. The output generated through this system, as demonstrated by using the technique of Grad-CAM over the image to show which regions of the image contribute the most to the decision made by the model, is displayed to the user through a simple desktop application which, also outputs whether the sample is pure or adulterated.

Keywords: Food adulteration; deep learning; computer vision; transfer learning; explainable AI (search for similar items in EconPapers)
Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://ijsrst.com/home/article/view/IJSRST26133229 Abstract page (text/html)
https://ijsrst.com/home/article/download/IJSRST26133229/IJSRST26133229 Full text (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:etm:ijsrst:v13:y2026:i3:id:1688

DOI: 10.32628/IJSRST26133229

Access Statistics for this article

More articles in International Journal of Scientific Research in Science and Technology from Technoscience Academy
Bibliographic data for series maintained by Pankaj Sharma ().