RESUMO
Recent years have witnessed the great advances of deep neural networks (DNNs) in light field (LF) image super-resolution (SR). However, existing DNN-based LF image SR methods are developed on a single fixed degradation (e.g., bicubic downsampling), and thus cannot be applied to super-resolve real LF images with diverse degradation. In this article, we propose a simple yet effective method for real-world LF image SR. In our method, a practical LF degradation model is developed to formulate the degradation process of real LF images. Then, a convolutional neural network is designed to incorporate the degradation prior into the SR process. By training on LF images using our formulated degradation, our network can learn to modulate different degradation while incorporating both spatial and angular information in LF images. Extensive experiments on both synthetically degraded and real-world LF images demonstrate the effectiveness of our method. Compared with existing state-of-the-art single and LF image SR methods, our method achieves superior SR performance under a wide range of degradation, and generalizes better to real LF images. Codes and models are available at https://yingqianwang.github.io/LF-DMnet/.
RESUMO
Introduction: Improvement of root architecture is crucial to increasing nutrient acquisition. Methods: Two pot experiments were conducted to investigate the effects of different concentrations of urea ammonium nitrate solution (UAN) and ammonium polyphosphate (APP) on lettuce root architecture and the relationship between roots and nitrogen (N) and phosphorus (P) absorption. Results: The results showed that lettuce yield, quality, and root architecture were superior in the APP4 treatment compared to other P fertilizer treatments. The N480 treatment (480 mg N kg-1 UAN) significantly outperformed other N treatments in terms of root length, root surface area, and root volume. There were significant quantitative relationships between root architecture indices and crop uptake of N and P. The relationships between P uptake and root length and root surface area followed power functions. Crop N uptake was significantly linearly related to the length of fine roots with a diameter of <0.5 mm. Conclusion and discussion: The length of fine roots played a more prominent role in promoting N absorption, while overall root size was more important for P absorption. APP has a threshold of 9.3 mg P kg-1 for stimulating the root system. Above this threshold, a rapid increase in root absorption of P. UAN can promote extensive growth of fine roots with a diameter less than 0.5 mm. Applying appropriate rates of APP and limiting UAN application to less than 400 mg N kg-1 can improve root architecture to enhance N and P absorption by lettuce. These results highlight a new possibility to improve nutrients use efficiency while maintaining high yields.