Scribd
Upload
18 views4 pages

Fruit Classification Project Documentation

The Fruit Classification Project aims to categorize fruit images (Apple, Banana, Orange) using machine learning models, achieving high accuracy with Decision Tree and SVM at 95.83%. The project involved dataset preparation, feature extraction, model training, and evaluation using Python and tools like Jupyter Notebook. Future work includes automating feature extraction and expanding the dataset for real-time classification.

Uploaded by

mr.rizwan8182
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
18 views4 pages

Fruit Classification Project Documentation

The Fruit Classification Project aims to categorize fruit images (Apple, Banana, Orange) using machine learning models, achieving high accuracy with Decision Tree and SVM at 95.83%. The project involved dataset preparation, feature extraction, model training, and evaluation using Python and tools like Jupyter Notebook. Future work includes automating feature extraction and expanding the dataset for real-time classification.

Uploaded by

mr.rizwan8182
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Date of Submission:

19/05/2025

Project Title: Fruit Classification Project


Documentation
Course Title: CS-504
Course Code: Digital Image Processing(DIP)

Submitted To:
Sir Muhammad Ismail

Submitted By:
Muhammad Rizwan
AG No: 2021-ag-7823
M Hammad Akhtar
AG No: 2021-ag-7740
Zohaib Maqsood
AG No: 2021-ag-7831
Ahmad Iqbal
AG No: 2021-ag-7899

Department of Computer
Science
Project Title: Fruit Classification Project
Documentation
Objectives
To classify fruit images into different categories (Apple, Banana, Orange) using extracted features and
machine learning models, and to evaluate model performance.

Key Features
This project focuses on building a fruit classification system using a dataset of fruit images. Images were
collected, categorized, and manually labeled with specific features. Three machine learning models
(Decision Tree, K-Nearest Neighbors, and Support Vector Machine) were trained and evaluated.

Code Implementation
• The dataset consists of three categories: Apple, Banana, Orange.
• Each category contains 35-40 images.
• Images were renamed sequentially and organized in separate folders per category.
• Features labeled for each image:
• Is_Round
• Color_Red
• Is_Elongated
• Peel_Texture_Visible

• Labels are encoded as 1 (feature present), -1 (feature absent), 0 (uncertain).

5. Tools and Technologies Used


• Python (Pandas, scikit-learn)
• Jupyter Notebook / Google Colab
• Microsoft Excel (for dataset labeling)
• Image dataset collected manually or via web search

6. Methodology
Step 1: Dataset Preparation
• Images collected and placed in folders named after categories.
• Images renamed sequentially ([Link], [Link], ...).
• Created an Excel sheet with image names, categories, and manually labeled features.
Step 2: Feature Extraction
• Basic image features manually labeled based on visual characteristics.

1
Step 3: Model Training
• Loaded labeled dataset into Jupyter Notebook.
• Split data into training and testing sets.
• Trained three classifiers:
o Decision Tree
o K-Nearest Neighbors (KNN)
o Support Vector Machine (SVM)
Step 4: Model Evaluation
• Evaluated model accuracy on the test set.
• Compared results to select the best performing model.

7. Results
Model Accuracy (%)
Decision Tree 95.83
KNN 91.67
SVM 95.83
The Decision Tree and SVM models showed the highest accuracy of 95.83%.

8. Conclusion
The project successfully classified fruit images into three categories with high accuracy using simple
manually labeled features and traditional machine learning models. This demonstrates the effectiveness of
feature-based image classification and machine learning algorithms.

9. Future Work
• Automate feature extraction using advanced image processing or deep learning.
• Expand dataset with more fruit categories and images.
• Implement real-time fruit classification using a camera.

2
10. Code
Click here to view the Colab notebook.
[Link]
?usp=sharing

Output:

You might also like