Abstract:The industry responsible for providing fresh fruits to the market plays a crucial role in ensuring the quality and safety of the produce. Classifying and grading the freshness of fruits is essential to maintain high standards. The presence of harmful bacteria in fruits poses a significant threat to the global agriculture sector. Recently, there has been an increased awareness of the vulnerability of fruits to various diseases, adding economic pressure on the agriculture sector worldwide. To address this issue, the proposed system employs a combination of image processing algorithms and deep learning models to analyze the visual characteristics of fruits and accurately determine their freshness. Addressing the challenges associated with subjective visual judgment and time-consuming manual inspection, the research introduces a methodology termed Machine Learning Life Cycle (MLLC), comprising six distinct phases from data gathering to the test model. Utilizing a carefully curated dataset sourced from Kaggle, featuring fresh and rotten apples, oranges, and bananas, the research involves training and classification process diagrams and algorithm implementation covering image pre-processing and Convolutional Neural Network (CNN) based model training. Evaluation metrics, including classification reports and confusion matrices, reveal assessed performance across fruits, with bananas demonstrating the highest accuracy which is 85%. The research aims to develop a web-based deep learning system for fruit freshness detection but has a limitation in optimal feature segmentation using the Gray-Level Co-occurrence Matrix (GLCM) for fruit freshness detection. Challenges include selecting relevant features across diverse fruit images, potential overfitting with deep convolutional layers, and balancing model complexity with data availability. Future works should explore advanced feature extraction techniques, mitigate overfitting risks, and enhance user interface inclusivity through continuous optimization informed by feedback and accessibility testing.

Meta-Analysis of Machine Learning Methods for Fruit Quality Prediction

YOLO Models for Fresh Fruit Classification from Digital Videos

Using Machine Learning Approaches for Food Quality Detection

Facilitated machine learning for image-based fruit quality assessment

Enhancing Spectroscopy-Based Fruit Quality Control: A Knowledge-Guided Machine Learning Approach to Reduce Model Uncertainty

Prediction of Fruit Maturity, Quality, and Its Life Using Deep Learning Algorithms

A Systematic Review and Comparative Meta-analysis of Non-destructive Fruit Maturity Detection Techniques

A Review on Automated Detection and Assessment of Fruit Damage Using Machine Learning

Detection of citrus black spot disease and ripeness level in orange fruit using learning-to-augment incorporated deep networks

Robust deep learning method for fruit decay detection and plant identification: enhancing food security and quality control

Strawberry Ripeness Detection Using Deep Learning Models

An Improved CNN-based Neural Network Model for Fruit Sugar Level Detection

Classifying Healthy and Defective Fruits with a Multi-Input Architecture and CNN Models

An Artificial Intelligence Approach Toward Food Spoilage Detection and Analysis

A general purpose multi-fruit system for assessing the quality of fruits with the application of recurrent neural network

Apple Ripeness Identification Using Deep Learning

Research Progress of Machine Learning in Extending and Regulating the Shelf Life of Fruits and Vegetables

Grading Methods for Fruit Freshness Based on Deep Learning

GLCM-Based Feature Extraction and CNN Approach for Fruit Freshness Detection

A Deep Learning Approach to Improving Spectral Analysis of Fruit Quality under Interseason Variation

Review: The evolution of chemometrics coupled with near infrared spectroscopy for fruit quality evaluation. II. The rise of convolutional neural networks