Visual Transformer for Soil Classification

TL;DR

This paper demonstrates that Visual Transformers significantly improve soil classification accuracy over traditional models, achieving over 93% accuracy on a dataset of four soil types, thus offering a promising approach for agricultural applications.

Contribution

The study introduces the application of Visual Transformers for soil classification and compares their performance with other deep learning and machine learning models, highlighting their superior accuracy.

Findings

Visual Transformers outperform other models by at least 2% in accuracy.

Achieved 98.13% training and 93.62% testing accuracy.

Validated the effectiveness of Visual Transformers for soil type classification.

Abstract

Our food security is built on the foundation of soil. Farmers would be unable to feed us with fiber, food, and fuel if the soils were not healthy. Accurately predicting the type of soil helps in planning the usage of the soil and thus increasing productivity. This research employs state-of-the-art Visual Transformers and also compares performance with different models such as SVM, Alexnet, Resnet, and CNN. Furthermore, this study also focuses on differentiating different Visual Transformers architectures. For the classification of soil type, the dataset consists of 4 different types of soil samples such as alluvial, red, black, and clay. The Visual Transformer model outperforms other models in terms of both test and train accuracies by attaining 98.13% on training and 93.62% while testing. The performance of the Visual Transformer exceeds the performance of other models by at least 2%. Hence, the novel Visual Transformers can be used for Computer Vision tasks including Soil Classification.