Enhancing Papaya Leaf Disease Detection with CNN and Transfer Learning Fusion for Precise Disease Diagnosis

Papaya cultivation plays a vital role in global agriculture, providing a crucial source of nutrition and economic stability. However, the threat of diseases poses a significant challenge to papaya plant health. To address this challenge, we proposed an innovative approach for enhancing the detection of papaya leaf diseases using Convolutional Neural Networks (CNNs) and transfer learning fusion. Our proposed framework leverages the strengths of CNNs, known for their ability to extract intricate features from images, and the transfer learning approach combines these techniques to create a robust and efficient model for accurate papaya leaf disease diagnosis. Transfer learning approach with two pre-trained models: VGG-16 and ResNet-50 are taken to detect papaya leaf diseases. The training process involves utilizing a pre-trained model and fine-tuning it on a specific papaya leaf disease dataset. For experimentation, five different classes of papaya leaves such as Fresh Papaya Leaf, Papaya Black Spot, Papaya Leaf Curl, Papaya Ringspot, and Powdery Mildew of Papaya are considered. Experimental results demonstrate the effectiveness of our proposed approach in accurately identifying and classifying various papaya leaf diseases. The proposed model has received an accuracy of 99.79% using ResetNet-50 followed by CNN whose accuracy was 99.42% and VGG-16, whose accuracy was 98.25%. This fusion approach aims to create a robust and efficient model capable of distinguishing subtle patterns indicative of various papaya leaf diseases. This method not only improves the efficiency of disease detection but also demonstrates the adaptability of the model to various disease patterns. Our study contributes to the advancement of automated papaya leaf disease detection, providing a reliable and precise tool for early diagnosis. Overall, our approach offers a promising solution for improving papaya cultivation and ensuring a sustainable future for global agriculture.