An Enhanced Computer Vision and Clustering Approach for Agricultural Produce Classification

Agricultural supply chains face specific challenges in ensuring that produce is efficiently categorized, processed, and transported from farmers to end consumers. One of the key difficulties lies in accurately classifying agricultural products based on their quality and characteristics, which directly impacts decision-making processes regarding their handling, storage, and transportation. To address the research gap, this research paper proposes an advanced system that employs a K-means classifier integrated with Computer Vision techniques. The K-means algorithm, known for its ability to group data points into distinct clusters, is used here to classify agricultural produce by analysing visual features derived through edge detection and image augmentation. This proposed method enables the system to effectively identify and categorize produce into quality-based clusters, which allows for more targeted and efficient handling procedures. By automating the classification process, this proposed solution significantly reduces human error and its subjectivity in sorting agricultural goods. The system demonstrated a high performance with an overall accuracy of 99.03%, and for certain produce categories, it achieved optimal levels of precision 99.12%, recall 99.03%, and F1-score 99.035% These metrics highlight the system's robustness in differentiating between classes of agricultural produce, ensuring that each product is categorized correctly according to pre-defined standards. The Kmeans classifier, combined with the use of Computer Vision techniques, facilitates faster and more reliable decision-making processes, which in turn reduces delays in sorting and transportation. This enhanced classification method not only optimizes resource allocation but also minimizes waste by ensuring that produce is handled in the most efficient manner. Ultimately, the proposed system benefits all stakeholders in the agricultural supply chain, improving operational efficiency, reducing costs, and enhancing overall productivity.