Crop yield estimation is an important component in precision agriculture enabling farmers to optimize the resources and increase productivity through better decision-making. Traditional approaches such as statistical models and manual sampling are time-consuming and labor-intensive. Considering these issues in this paper, we propose a meta-learning enhanced self-supervised approach for crop yield images using UAV (unmanned ariel vehicle) imagery of crops. The proposed model also utilizes EfficientNet architecture to perform feature extraction in large-scale agriculture environments. Self-supervised learning is used to train the model on large-scale unlabeled data without manual annotations. Meta-learning helps to generalize the learning with minimal fine-tuning of model parameters making it suitable for diversified farming landscapes. The performance of the proposed approach is evaluated using different baseline methods. The results reveal that the proposed approach performs better in terms of computational efficiency and accuracy compared to existing methods improving crop yield prediction improving the efficiency of farming operations.