• Suren Kumar Sahu
• Monalisha Behera
• S.Gayatri Dora

In terms of economic viability and human existence, agriculture is a nation's backbone. We aim to prevent plant diseases to maintain high agricultural production efficiency. In order to improve the outcome without adding complexity, the suggested method will optimise the data from the resources that are now accessible. The Dense Convolution Neural Network (DCNN) is the neural network that is employed for classification. For training in this project, a pre-trained neural network model (densenet-121) imported from the keras library was employed. An input that results in activation may be subjected to a convolution by the straightforward application of a filter. A feature map, which shows the positions and intensity of a detected feature in an input such as an image, is produced by repeatedly applying an analogous filter to an input. The convolutional networks assist in automatically learning a large number of filters relevant to a training dataset in simultaneously. The country's economy benefits as a consequence of this algorithm's ability to identify plant illnesses effectively. The densenet-121 has been used to classify the 29 various illnesses for 7 plants (potato, tomato, maize, bell pepper, grape, apple, and cherry) using 34599 photos from the Huges DP Plant- Village dataset from Kaggle. In my research, the original picture is transformed into HSV colour form, after which the thresholding-generated masked image is delivered to the suggested model for training and classification, yielding an average accuracy 98.23% (theoretical). We obtained an average accuracy of 94.96% for 50 epochs with a learning rate of 0.002 when all classes of plant disease were provided to the model for training using the Google Colab platform (Tesla-T4 processor).

Metallic surface, VGG-16, Neural Networks, defect detection

"PLANT DISEASE CLASSIFICATION USING DENSENET-121 ARCHITECTURE AND FLASK", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 9, page no.a156-a163, September-2023, Available :http://www.jetir.org/papers/JETIR2309022.pdf

"PLANT DISEASE CLASSIFICATION USING DENSENET-121 ARCHITECTURE AND FLASK", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 9, page no. ppa156-a163, September-2023, Available at : http://www.jetir.org/papers/JETIR2309022.pdf