As an alternative to true isotropic 3D imaging, image super-resolution (SR) has been applied to reconstruct an isotropic 3D volume from multiple anisotropic scans. However, traditional SR methods struggle with inadequate performance, prolonged processing times, and the necessity for manual feature extraction. Motivated by the exceptional representational ability and automatic feature extraction of convolutional neural networks (CNNs), in this work, we present an end-to-end isotropic MRI reconstruction strategy based on deep learning. The proposed method is based on 3D convolutional neural networks (3D CNNs), which can effectively capture the 3D structural features of MRI volumes and accurately predict potential structure. In addition, the proposed method takes multiple orthogonal scans as input and thus enables the model to use more complementary information from different dimensions for precise inference. Experimental results show that the proposed algorithm achieves promising performance in terms of both quantitative and qualitative assessments. In addition, it can process a 3D volume with a size of 256 × 256 × 256 in less than 1 min with the support of an NVIDIA GeForce GTX 1080Ti GPU, which suggests that it is not only a quantitatively superior method but also a practical one.

The Tibetan Plateau (TP) is experiencing higher warming rates than elsewhere, which may affect regional vegetation growth. Particularly on the Eastern Tibetan Plateau (ETP), where the topography is diverse and rich in biodiversity, it is necessary to clarify the drivers of climate and topography on vegetation cover. In this research, we selected the Shaluli Mountains (SLLM) in the ETP as the study area, monitored the spatial and temporal dynamics of the regional vegetation cover using remote sensing methods, and quantified the drivers of vegetation change using Geodetector (GD). The results showed a decreasing trend in annual precipitation (PRE) (−2.4054 mm/year) and the Palmer Drought Severity Index (PDSI) (−0.1813/year) in the SLLM. Annual maximum temperature (TMX) on the spatial and temporal scales showed an overall increasing trend, and the regional climate tended to become warmer and drier. Since 2000, fractional vegetation cover (FVC) has shown a fluctuating upward trend, with an average value of 0.6710, and FVC has spatially shown a pattern of “low in the middle and high in the surroundings”. The areas with non-significant increases (p > 0.05) and significant increases (p < 0.05) in FVC accounted for 46.03% and 5.76% of the SLLM. Altitude (q = 0.3517) and TMX (q = 0.3158) were the main drivers of FVC changes. As altitude and TMX increased, FVC showed a trend of increasing and then decreasing. The results of this study help us to clarify the influence of climate and topography on the vegetation ecosystem of the ETP and provide a scientific basis for regional biodiversity conservation and sustainable development.

In recent years, vision transformers have demonstrated remarkable superiority to CNNs in single image super-resolution (SISR), yet their heavy computational and storage costs hinder practical deployment. In this work, we revisit CNN-based residual learning for SISR and identify a specific yet overlooked problem: the residual signal is often biased by mapping errors during model training, and this bias can propagate and accumulate through layers. To this end, we revisit the commonly used residual learning and feature fusion in SISR and propose an adaptive residual correction network (AdaRCN) in this work. First, we introduce a residual correction mechanism that adaptively compensates for the bias in the residual, which is utilized to ease error accumulation and improve mapping accuracy. On the other hand, we generalize the standard identity shortcut to a weighted channel concatenation followed by a 1 × 1 convolution, which is a more versatile strategy for adaptive feature fusion. Our AdaRCN is built entirely upon a naive CNN without complex architecture design and training strategies, thus ensuring efficient inference and parallelization. Extensive experiments verify the benefits and effectiveness of residual correction and adaptive feature fusion in improving the representational capability of our model, enabling it to achieve impressive performance comparable to advanced SISR models with moderate overhead.

Objective: This paper designs a FACA model for deep mineral prediction in actual mining areas. Method: The spatial distribution of geochemical anomalies was consistent with the concentration--area (C-A), this paper established a C–A model for geochemical anomaly extraction. Based on mineral resources formed by multiple element combinations, factor analysis (FA) was used to obtain the different combinations and comprehensive information of elements. On this basis, a new FACA model was designed for mineral prediction using the FA and C–A. Results: The proposed FACA model was applied to mineral prediction in the Jiguanzui copper-gold mining area in China. The elements in the study area were divided into four combinations. The thresholds of single element and element combination anomalies were identified. Through diagnostic testing, the abnormal distributions of geochemical elements were consistent with their theoretical distributions, and the comprehensive abnormal distribution area of elements was consistent with the distribution of the actual ore bodies, demonstrating that the designed FACA algorithm of this paper was reasonable. Conclusions: A new blind ore body in the study area is predicted by using FACA model, positioned at a depth ranging from approximately 1120m to 1150m below ground, between drill holes ZK02618 and KZK23. These findings hold significant implications for mineral exploration efforts.