Real-Time Fruit Classification using Deep Learning via Web Interface for Automated Harvesting Applications

Real-Time Fruit Classification using Deep Learning via Web Interface for Automated Harvesting Applications

IEEE BASE PAPER TITLE:

Automated AI-Powered Fruit Identification using Convolutional Neural Network

IEEE BASE PAPER ABSTRACT:

Clever system that can look at pictures of fruits and figure out what kind of fruit each picture shows. AI algorithms like deep learning, which is like giving the Machine learning model a crash course in fruit recognition. This method teaches the ML Model we clearly explained the Model using application’s agent using PEAS and the application’s task environment using the 6 dimensions by showing it tons of fruit images, so over time, it gets really good at spotting the differences and similarities between, say, a banana and a grape. We also used another technique called pattern recognition, which helps the computer pay attention to specific details like the fruit’s color, shape, size and texture. Overcoming multiple obstacles in order to automatically identify the type of fruit from the picture. The variety of images is influencing the color, texture, and shape of many different types of fruits. When it came to fruit picture detection, Convolutional Neural Network (CNN) Algorithm outperformed standard support-vector-machine-based approaches using handcrafted features in terms of accuracy also, it is a lot faster to implement for new fruits. By integrating deep learning and pattern recognition techniques such as Convolutional Neural Network Algorithm we have got the accuracy of 84%, our system efficiently identified different fruit types from images, demonstrating the power and effectiveness of our methods. The goal of our project is to create a tool that can quickly and correctly identify many types of fruits in photos, which could be useful for things like sorting fruits in a grocery store or helping people learn about different fruits by using Convolutional Neural Network Algorithm. This is not just about teaching a computer to recognize fruits; it is about making technology that can understand and interact with the world in a way that is helpful to us.

PROJECT OUTPUT VIDEO:

ALGORITHM / MODEL USED:

InceptionV3, MobileNet.

OUR PROPOSED PROJECT ABSTRACT:

The growing demand for automation in agriculture and food processing industries has fueled the development of intelligent fruit classification systems. This project, titled “Real-Time Fruit Classification using Deep Learning via Web Interface for Automated Harvesting Applications” presents an end-to-end deep learning-based solution for identifying and classifying fruit types with high precision and efficiency. The system is implemented using Python as the backend language and a web interface built with HTML, CSS, JavaScript, and the Flask web framework to enable real-time, user-friendly interaction.

The core of the system leverages two powerful Convolutional Neural Network (CNN) architectures: InceptionV3 and MobileNet, trained on a comprehensive fruit image dataset comprising 100 distinct fruit classes. These classes include diverse varieties such as Apple Braeburn, Avocado ripe, Banana Red, Cantaloupe, Dragon Fruit (Pitahaya Red), Mango, Pear Abate, Pineapple Mini, Raspberry, Strawberry Wedge, and Watermelon, among others. Total dataset images are 68927. The use of this large-scale dataset ensures the system’s robustness in handling a wide range of fruit types under varying visual conditions.

The InceptionV3 model achieved a training accuracy of 99.4% and a test accuracy of 93.4%, while the lightweight MobileNet architecture outperformed it with a training accuracy of 99.7% and a test accuracy of 97.6%, making it highly suitable for deployment in real-time applications, especially on devices with limited computational resources. Extensive performance evaluations were conducted, including graphical representations of training and validation loss and accuracy metrics for both models. A comparative model accuracy chart is also provided to visually assess and contrast the performance of InceptionV3 and MobileNet.

By integrating these advanced models into a real-time, web-accessible interface, the proposed system offers a scalable solution for automated harvesting applications, where instant and accurate fruit recognition is crucial. This system not only facilitates automated fruit sorting and quality control in agricultural domains but also serves as a foundation for further innovations in smart farming and AI-driven produce management.

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS: 

• System : Pentium i3 Processor.
• Hard Disk : 20 GB.
• Monitor : 15’’ LED.
• Input Devices : Keyboard, Mouse.
• Ram : 8 GB.

SOFTWARE REQUIREMENTS: 

• Operating System : Windows 10 / 11.
• Coding Language : Python 3.12.0.
• Web Framework : Flask.
• Frontend : HTML, CSS, JavaScript.

REFERENCE:

Sai Tejasree K; Surendran R; Rajakumar B; Gomathi R. M, “Automated AI-Powered Fruit Identification Using Convolutional Neural Network”, 2025 International Conference on Multi-Agent Systems for Collaborative Intelligence (ICMSCI), IEEE XPLORE 2025.

👉CLICK HERE TO BUY THIS PROJECT “Real-Time Fruit Classification using Deep Learning via Web Interface for Automated Harvesting Applications” SOURCE CODE👈

Frequently Asked Questions (FAQ’s) & Answers:

Q1: What is the purpose of this project?

The purpose of this project is to automatically classify fruits into 100 different categories using deep learning models. The system provides a web-based interface that allows users to upload fruit images and receive real-time classification results.

Q2: Which technologies are used in this project?

The backend is developed using Python and Flask. The frontend is designed using HTML, CSS, and JavaScript. Deep learning models such as InceptionV3 and MobileNet are used for image classification.

Q3: How many fruit types can this system classify?

The system can classify 100 distinct types of fruits, including apples, bananas, grapes, berries, melons, and many exotic fruits.

Q4: What is the accuracy of the models used?

• InceptionV3: 99.4% training accuracy and 93.4% test accuracy. • MobileNet: 99.7% training accuracy and 97.6% test accuracy.

Q5: Can users switch between different models?

Yes, the system allows switching between InceptionV3 and MobileNet models based on user preference or performance requirements.

Q6: How does the system handle invalid or unknown inputs?

If a non-fruit image or unsupported file format is uploaded, the system provides the name which is a closest possible match.

Q7: Is the system capable of real-time classification?

Yes, the system is optimized for real-time predictions, with average response times under 3 seconds for both models.

Q8: Can this system be used on mobile devices?

No. We have developed only the Web Application. Currently, the system is web-based and can be accessed through a mobile in future. A mobile application version can be considered in future development.

Q9: What is the dataset used in this project?

The dataset contains high-quality images of 100 fruit categories. Each category includes multiple images captured under various lighting and background conditions to ensure diversity and robustness.

Q10: What are the practical applications of this project?

This system can be used in agriculture (automated sorting and quality control), supermarkets (inventory systems), educational tools (learning about fruits), and mobile commerce platforms for fruit identification.

Q11: Is the source code customizable for more fruit classes?

Yes, the codebase is modular and can be extended to include additional fruit classes by retraining the models with new data. The model should be trained in the backend to identify new classes of fruits.

Q12: Is internet required to use this system?

Yes, since it is a web-based application, an internet connection is required to access and use the system online. However, it can also be deployed locally in offline environments with minimal modifications.