Abstract

Food quality assessment is critical in ensuring safety, freshness, and nutritional value in the food supply chain. Traditional manual inspection methods are often subjective, time-consuming, and error-prone, necessitating the development of automated, reliable systems. Existing image processing-based food quality systems lack accuracy, real-time operability, or efficient integration into embedded hardware. They also struggle with variable lighting conditions and different types of food textures, leading to inconsistent results. This study proposes a high-quality embedded system that uses histogram-based image analysis to assess food quality. The system integrates a Raspberry Pi 4 with a high-resolution camera module to capture food images. The images undergo preprocessing steps including RGB to grayscale conversion, histogram equalization, and noise reduction. Feature extraction is then performed using histogram intensity distributions, which are analyzed for quality grading. The histogram data is classified using a trained SVM model implemented in Python and OpenCV. Experimental results show that the proposed system achieves 93.8% accuracy in food quality classification across diverse food items such as fruits and vegetables. Compared to existing methods, our approach demonstrated higher precision, better real-time performance, and lower hardware costs. The system is lightweight, scalable, and suitable for deployment in farms, markets, or homes.

Authors

B. Guruprakash¹, Babasaheb Dnyandeo Patil²
Sethu Institute of Technology, India¹, Bharati Vidyapeeth, India²

Keywords

Food Quality Assessment, Embedded System, Histogram Analysis, Image Processing, SVM Classification