Papaya is an important fruit due to its economic importance and nutritional benefits. There is a high amount loss of papaya in farms due to its diseases. The process of papaya diseases early detection is crucial to protect the yield and reduce losses. In this chapter, VGG16-DNN model is proposed to recognize healthy and unhealthy papaya leaves. The unhealthy leaves splits into five diseases: Curl, Mealybug, Mite, Mosaic and Ring Spot. It utilizes the pre-trained VGG16 model as feature extraction after removing the top layer and adding three fully connected layers to complete the classification process. VGG16 consists of five convolutional blocks with different number of convolutional layers (CNNs) and filters. Each block is followed by max-pooling layer to reduce the size of feature map. The dataset provides 6,658 leaves splitted into 20% for training set and 20% for testing set. The proposed model achieves classification metrics 97% across accuracy, precision, recall, and f1-score. The experimental results prove the superiority of the proposed model compared to ResNet50 and Xception. Additionally, it has an acceptable inference time that makes it suitable for real-time applications.