EA-CNN: Enhanced attention-CNN with explainable AI for fruit and vegetable classification
2024

Enhanced Attention-CNN for Classifying Fruits and Vegetables

Sample size: 90000 publication 10 minutes Evidence: high

Author Information

Author(s): Zeshan Aslam Khan, Waqar Muhammad, Khalid Mehmood Cheema, Bakar Mahmood Ali Abu, Ain Quratul Ain, Chaudhary Naveed Ishtiaq, Alshehri Abdullah, Alshamrani Sultan S., Zahoor Raja Muhammad Asif

Primary Institution: National Yunlin University of Science and Technology

Hypothesis

Can an enhanced attention-CNN model improve the classification accuracy of fruits and vegetables using visual features?

Conclusion

The proposed EA-CNN model achieved a classification accuracy of 98.1% on the Fruit-360 dataset and 96% on the Fruit Recognition dataset, outperforming existing models.

Supporting Evidence

  • The EA-CNN model outperformed baseline models in terms of accuracy and computational cost.
  • The model achieved a test accuracy of 98.1% on the Fruit-360 dataset.
  • The proposed model demonstrated substantial performance on the Fruit Recognition dataset with a generalized accuracy of 96%.
  • Utilizing the entire Fruit-360 database significantly enhanced the model's classification capabilities.

Takeaway

This study created a smart computer program that can tell different fruits and vegetables apart by looking at their pictures, and it does this really well!

Methodology

The study used a simplified CNN architecture with an attention mechanism and a customized pooling technique to classify fruits and vegetables from the Fruit-360 dataset.

Limitations

The study may be limited by the quality and diversity of the datasets used for training and testing.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1016/j.heliyon.2024.e40820

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication