I-VIEW: AN IMAGE PROCESSING AND MACHINE LEARNING APPROACH FOR ACCURATE IDENTIFICATION OF SOYBEAN (Glycine max L.) CROP AND WEEDS
No Thumbnail Available
Views
...
Downloads
...
Date
2024
Authors
Ronny Angel B. Pastrana
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In recent years, the classification of soybean plants and weeds has gained traction due to the increasing need for automated agricultural solutions. To address this, the study introduces I-VIEW, also know as Intelligent View, a deep learning-based approach using Convolutional Neutral Networks (CNNs) to classify soybean plants varieties, including Black, Brown Canada, Clsoy 1, Clsoy 2, Collection 1, Collection 2 and Tiwala 10 Soybean, incorporating advanced image processing techniques to improve accuracy and RGB image processing to prepare the dataset for training. The deep learning models for I-VIEW were developed using Python, a versatile and widely used programming language known for its robust ecosystem of machine learning libraries such as TensorFlow and Keras. the evaluation was conducted at three levels: overall classification accuracy on testing dataset, validation dataset, and real-world testing. The testing dataset achieved a perfect accuracy off 100% with no errors among 9000 images; the validation dataset attained accuracy rate of 94.7% with a 5.3% error rate. In real-world scenarios, the accuracy rate is 91% and 9% error rate. The proposed method outperforms previous relevant works and provides a reliable approach to identifying soybean plants and weeds.