Skip to content

Latest commit

 

History

History
41 lines (33 loc) · 1.31 KB

README.md

File metadata and controls

41 lines (33 loc) · 1.31 KB
Raw

Seminar and Technical Communication - Semester 5

A Fruit Classification System using CNN.

Table of Contents

  1. Abstract
  2. Motivation
  3. Modules Used
  4. Proposed Architecture
  5. Conclusion

Abstract

Fruit classification is a challenging process which has always been done manually. We can try to create an automatic fruit categorization system using machine learning. Fruits have an uneven shape, a wide range of sizes, and a diversity of colours, making it difficult to identify. However, developing an efficient classification system will counter these difficulties. Such systems can be further improved to do the task of fruit grading along with fruit classification.

Motivation

  • Smart Agriculture
  • Reducing labor cost
  • Time saving approach
  • Helping producers and businesses to grow

Modules Used

  • NumPy
  • Pandas
  • Matplotlib
  • Keras
  • Tensorflow

Proposed Architecture

CNN-architecture

Figure: System Architecture

Conclusion

  • loss: 1.2458
  • accuracy: 0.8833
  • System can be further improved to perform the task of fruit grading and fruit disease detection along with fruit classification.