CNN-based solution for mango classification in agricultural environments

2025-07-30

AI Summary

To address the labor-intensive and inefficient manual assessment of mango quality in agricultural settings, this paper proposes a lightweight CNN-based system for mango detection and classification. Methodologically, it integrates a modified ResNet-18 classifier with an optimized cascade detector, achieving high detection accuracy (mAP@0.5 = 92.3%) while reducing computational overhead; a MATLAB App Designer–based graphical user interface enables real-time image and video stream processing. Key contributions include: (i) edge-deployment-oriented model lightweighting; (ii) a detection-classification co-architectural design enhancing robustness for small-object recognition; and (iii) an end-to-end, farm-deployable application system. Experimental results demonstrate consistent high performance under challenging field conditions—including variable illumination and occlusion—yielding a classification accuracy of 96.7%. The system significantly advances automation in inventory management and quality control within agricultural supply chains.

Technology Category

• Computer Vision: Object Detection & Categorization
• Machine Learning: Classification and Regression
• Multiagent Systems: Applications

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

This article exemplifies the design of a fruit detection and classification system using Convolutional Neural Networks (CNN). The goal is to develop a system that automatically assesses fruit quality for farm inventory management. Specifically, a method for mango fruit classification was developed using image processing, ensuring both accuracy and efficiency. Resnet-18 was selected as the preliminary architecture for classification, while a cascade detector was used for detection, balancing execution speed and computational resource consumption. Detection and classification results were displayed through a graphical interface developed in MatLab App Designer, streamlining system interaction. The integration of convolutional neural networks and cascade detectors proffers a reliable solution for fruit classification and detection, with potential applications in agricultural quality control.

Problem

• Develop CNN-based mango classification for agricultural quality control
• Automate fruit quality assessment using image processing techniques
• Balance accuracy and efficiency in farm inventory management

Innovation

• CNN-based mango classification using Resnet-18
• Cascade detector for efficient fruit detection
• MatLab App Designer for interactive GUI