Abstract: During the conventional agricultural era, farmers are not keen on increasing production due to the lack of effective approaches for diagnosing diseases in different crops. Early identification of crop diseases is vital as it significantly influences the growth of plant species. While a variety of Machine Learning (ML) models have been employed to organize and identify agricultural diseases, recent advancements in Deep Learning (DL) provide substantial promise for enhancing precision in this domain. The suggested approach accurately and efficiently detects crop disease symptoms by using a neural network based on convolution (CNN). Model performance is evaluated using a variety of efficiency signs, which show how effective it is in early disease identification. The report fills research gaps for reliable disease detection approaches and offers a thorough investigation of deep learning frameworks for crop disease visualization. The suggested convolutional neural network approach seeks to transform the plant leaf diseases identification, including those that do not yet exhibit symptoms. Expanding on prior work emphasizing the significance of plant disease identification, this research introduces an advanced solution. With the support of an automated system and the deep learning algorithm Convolutional Neural Network (CNN), people can identify illnesses with cell phones. Furthermore, a stunning 99.81% classification accuracy is achieved by integrating DenseNet-121 into the framework, demonstrating its superiority over other models. This method has the potential to transform crop disease detection and boost food security and agricultural output.
Keywords: Convolutional Neural network, DenseNet- 121, Alternaria solani, Phytophthora infestans, Machine Learning, early bright, late blight.
Cite:
Prof. Himanshu V. Taiwade, Rakesh Nagrikar, Diksha Narekar, Prashik Nagrale, "MONITORING PLANT DISEASES USING A DEEP LEARNING – BASED APPROACH", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 13, no. 1, 2024, Crossref https://doi.org/10.17148/IJARCCE.2024.13116.
|
DOI:
10.17148/IJARCCE.2024.13116
