Authors : Saranya P.; Kalaiselvi K.; Kaviyazhini K.; Prithisha G.; Thirumalai Selvi M.

Volume/Issue : Volume 11 - 2026, Issue 4 - April

Google Scholar : https://tinyurl.com/5akz333v

Scribd : https://tinyurl.com/y7ey32ju

DOI : https://doi.org/10.38124/ijisrt/26apr915

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Fruit quality assessment plays a crucial role in ensuring food safety, reducing post-harvest losses, and maintaining supply chain efficiency. Traditional methods rely on manual inspection or basic image processing techniques, which are often subjective, time-consuming, and inaccurate under varying environmental conditions. This paper proposes a fruit quality assessment and freshness score prediction system using deep learning techniques. The system combines image classification and segmentation approaches to evaluate fruit conditions. It first verifies whether the input image contains an apple, and if not, returns an “object not found” response. For valid apple inputs, the system identifies defective regions such as rot and cracks and computes a quantitative freshness score based on the proportion of defective areas relative to the total fruit surface. Experimental results demonstrate high accuracy and improved reliability compared to traditional methods. The proposed system provides a scalable and effective solution for apple quality evaluation.

Keywords : Deep Learning, Fruit Quality Assessment, Freshness Score, U-Net; Mobile NetV, Image Classification, Semantic Segmentation Computer Vision.

References :

1. Chen, X., Li, H., & Wang, Y. (2026). ML Models for Predicting Post-Harvest Apple Quality. Journal of Food Quality and Preservation, 12(3), 45–59.
2. Kumar, R., Singh, P., & Verma, S. (2025). AI-driven Spoilage Prediction for Apples. International Journal of Food Science and Technology,60(7), 1123–1135.
3. Li, J., & Zhang, Q. (2025). Non-invasive Detection of Fruit Spoilage Using Hyperspectral Imaging. Journal of Agricultural Engineering, 58(4), 210–222.
4. Sharma, A., Patel, R., & Gupta, N. (2024). Machine Learning-based Early Detection of Apple Spoilage. Computers and Electronics in Agriculture, 203, 107–119.
5. Singh, A., & Rao, K. (2026). Computer Vision for Fruit Freshness Assessment. Journal of Food Engineering and Technology, 14(2), 88–102.
6. Fruit-360 Dataset, “Images of fruits with fresh and spoil labels for classification,” Kaggle, 2020, https://www.kaggle.com/datasets/ulnnproject/food-freshness-dataset