• Muhammad Harris Hashmi
• Bhanu Priya Mahant
• Muhammad Salik Salam
• Muhammad Ahsan

Plant diseases (PD) have a detrimental effect on agricultural output, quality, economics, and quantity. To detect such diseases, the life cycle of plants should be properly monitored from the beginning. It is necessary to automate the disease-detecting equipment to expedite this process. Developing a disease detection system should take advantage of deep learning (DL) techniques. No environmental issues had hindered agriculture until recently when several plant leaf diseases began to reduce harvests. It is difficult to detect PD early, which is only one of the many problems that producers face. To solve this problem, scientists created the 'PDConv model, an automated PD categorization system built on a deep convolution neural network (Deep CNN). In this four-step technique, we preprocessed the data, added it to it, extracted features, and classified the data. In this study, we classify data and extract features using a Deep CNN, the layered architecture of which is briefly discussed. The complete system was implemented with the help of the suggested method utilizing python programming language and the Jupyter notebook environment. Using data from the FGVC7 dataset, researchers can create deep learning (DL) PD classification system that is both accurate and efficient. When compared to CNN, the PDConv model performed exceptionally well. A CNN model may be trained with only the segmented and annotated parts of images rather than with whole images. After being trained on the preexisting CNN model, the PDConv model's results on independent data sources increased from 61.91% to 95.46%. In addition, 82% of the test data set showed an increase in self-assessment confidence, indicating strong development in this area. Proof that CNNs may be used effectively for identifying PD.

Plant Disease Detection, Image Segmentation, Classification, Deep Learning, Deep CNN, FGVC7 Dataset.

"A Deep Learning-Based PDConv Model for Plant Disease Detection, Classification and Prediction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 1, page no.d9-d14, January-2023, Available :http://www.jetir.org/papers/JETIR2301303.pdf

"A Deep Learning-Based PDConv Model for Plant Disease Detection, Classification and Prediction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 1, page no. ppd9-d14, January-2023, Available at : http://www.jetir.org/papers/JETIR2301303.pdf