AGRICULTURAL SEED IMAGE CLASSIFICATION USING CNN

Dr. A. Mary Sowjanya; Mrs. Varsha Kumari; Malireddi Sai Lahari

doi:10.5281/zenodo.22639518

Abstract

The rapid growth of the agricultural supply chain has increased the need for fast and reliable identification of seed
varieties, with manual visual inspection being one of the most significant bottlenecks in seed quality assessment.
Different seed varieties can appear visually similar, and traditional inspection methods often struggle to provide
consistent, scalable classification. This paper presents a Deep Learning-based Agricultural Seed Image
Classification System that analyzes seed images and classifies them into one of nine varieties: Black Gram,
Chickpea, Corn, Moong Bean, Pinto Bean, Rice, Soyabean, Wheat, and White Kidney Bean. The dataset of 900
labelled images is preprocessed through resizing, normalization, and data augmentation (rotation, shifting,
shearing, zooming, and horizontal flipping), and a Convolutional Neural Network based on the AlexNet
architecture is trained and evaluated. The experimental results show that the CNN model achieves a validation
accuracy of 94.00%, with precision of 0.9636, recall of 0.9629, F1-Score of 0.9635, and AUC-ROC of 0.9812.

Keywords

Seed Classification, Convolutional Neural Network, Deep Learning, AlexNet, Image Processing, Data Augmentation, Precision Agriculture.

References

[1] A. Krizhevsky, I. Sutskever, and G. E. Hinton, "ImageNet Classification with Deep Convolutional
Neural Networks," Advances in Neural Information Processing Systems (NIPS), vol. 25, pp. 1097-
1105, 2012.
[2] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning, MIT Press, 2016.
3) [3] F. Chollet, Deep Learning with Python, 2nd ed., Manning Publications, 2021.
4) [4] R. C. Gonzalez and R. E. Woods, Digital Image Processing, 4th ed.,
Pearson Education, 2018.
5) [5] C. M. Bishop, Pattern Recognition and Machine Learning, Springer,
2006.
6) [6] TensorFlow Developers, TensorFlow Documentation, https://www.tensorflow.org
7) [7] Keras Team, Keras Documentation, https://keras.io
8) [8] OpenCV Team, OpenCV Documentation, https://opencv.org
9) [9] NumPy Documentation, Scientific Computing with Python, https://numpy.org
10) [10] Scikit-learn Developers, Scikit-learn Documentation, https://scikit-learn.org
11) [11] Pandas Documentation, Python Data Analysis Library, https://pandas.pydata.org
12) [12] Matplotlib Documentation, Visualization with Python, https://matplotlib.org
13) [13] Seaborn Documentation, Statistical Data Visualization, https://seaborn.pydata.org