Sunflower leaf disease significantly cause damage to the yield and the quality of crop, which emphasize the need for accurate leaf disease detection and classification. This research examine the integration of both processing and an advanced Deep Learning (DL) classification technique for the prediction of disease in the leaf of the sunflower. The image of Sunflower Fruits and Leaves datasets are inserted for preprocessing based on Contrast Limited Adaptive Histogram Equalization (CLAHE) technique to enhance the image brightness and provide clear prediction of disease in the leaf image. The preprocessed image is segmented based on Fuzzy C-means (FCM) clustering, an effective segmentation technique that clusters multiple data with the different relationship. Local Binary Pattern (LBP) utilized for extracting the texture of the images and enables improved differentiation of normal and infectious leaf. A Contemporary Architecture for Convolutional Neural Networks (ConvNext) employed for image classifier of sunflower leaf for attaining enhanced accuracy in leaf disease prediction. The proposed model is implemented using Python software and evaluated based on performance metrics, achieving an accuracy of 93%, with precision, recall, and F1-score reaching 96%, effectively predicting disease in sunflower leaves. Finally, the proposed methodologies improves the prediction level and contribute to early prediction of the disease which leads to reduce the losses of the crop.