With the development of deep learning techniques in the field of remote sensing change detection, many change detection algorithms based on convolutional neural networks (CNNs) and nonlocal self-attention (NLSA) mechanisms have been widely used and have obtained good detection accuracy. However, these methods mainly extract semantic features on images from different periods without taking into account the temporal dependence between these features. This will lead to more “pseudo-change” in complex scenes. In this paper, we propose a network architecture named UVACD for bitemporal image change detection. The network combines a CNNs extraction backbone for extracting high-level semantic information with a visual transformer. Here, visual transformer constructs change intensity tokens to complete the temporal information interaction and suppress irrelevant information weights to help extract more distinguishable change features. Our network is validated and tested on both the LEVIR-CD and WHU datasets. For the LEVIR-CD dataset, we achieve an intersection over union (IoU) of 0.8398 and an F1 score of 0.9130. For the WHU dataset, we achieve an IoU of 0.8664 and an F1 score of 0.9284. The experimental results show that the proposed method outperforms some previous state of the art change detection methods.

The sensing depth of passive microwave remote sensing is a significant factor in quantitative frozen soil studies. In this paper, a microwave radiation response depth (MRRD) was proposed to describe the source of the main signals of passive microwave remote sensing. The main goal of this research was to develop a simple and accurate parameterized model for estimating the MRRD of frozen soil. A theoretical model was introduced first to describe the emission characteristics of a three-layer case, which incorporates multiple reflections at the two boundaries. Based on radiative transfer theory, the total emission of the three layers was calculated. A sensitivity analysis was then performed to demonstrate the effects of soil properties and frequency on the MRRD based on a simulation database comprising a wide range of soil characteristics and frequencies. Sensitivity analysis indicated that soil temperature, soil texture, and frequencies are three of the primary variables affecting MRRD, and a definite empirical relationship existed between the three parameters and the MRRD. Thus, a parameterized model for estimating MRRD was developed based on the sensitivity analysis results. A controlled field experiment using a truck-mounted multi-frequency microwave radiometer (TMMR) was designed and performed to validate the emission model of the soil freeze–thaw cycle and the parameterized model of MRRD developed in this work. The results indicated that the developed parameterized model offers a relatively accurate and simple way of estimating the MRRD. The total root mean square error (RMSE) between the calculated and measured MRRD of frozen loam soil was approximately 0.5 cm for the TMMR’s four frequencies.

Remote sensing change detection (CD) identifies changes in each pixel of certain classes of interest from a set of aligned image pairs. It is challenging to accurately identify natural changes in feature categories due to unstructured and temporal changes. This research proposed an effective bi-temporal remote sensing CD comprising an encoder that could extract multiscale features, a decoder that focused on semantic alignment between temporal features, and a classification head. In the decoder, we constructed a new convolutional attention structure based on pre-generation of depthwise-separable change-salient maps (PDACN) that could reduce the attention of the network on unchanged regions and thus reduce the potential pseudo-variation in the data sources caused by semantic differences in illumination and subtle alignment differences. To demonstrate the effectiveness of the PDA attention structure, we designed a lightweight network structure for encoders under both convolution-based and transformer architectures. The experiments were conducted on a single-building CD dataset (LEVIR-CD) and a more complex multivariate change type dataset (SYSU-CD). The results showed that our PDA attention structure generated more discriminative change variance information while the entire network model obtained the best performance results with the same level of network model parameters in the transformer architecture. For LEVIR-CD, we achieved an intersection over union (IoU) of 0.8492 and an F1 score of 0.9185. For SYSU-CD, we obtained an IoU of 0.7028 and an F1 score of 0.8255. The experimental results showed that the method proposed in this paper was superior to some current state-of-the-art CD methods.

The topographic effect is one of the essential factors affecting microwave radiation in mountainous regions. This effect's correction can improve the inversion accuracy of key surface parameters. Here, a new terrain correction model in microwave pixel scale is proposed by regression. Using the rugosity (RU) measure the relief of land surface, and analyze its correlation with the brightness temperature difference between the undulating surface and flat surface ( $\Delta TB_{\varphi_ {Ri}}$ ). The topographic effects are simulated by the mountain microwave radiation model. Through the regression analysis of the topographic effects with RU in different regions, we find that: first, RU has a linear relationship with the average $\Delta TB_{\varphi_ {Ri}}$ for different azimuths of the pixel ( $\overline{\Delta TB}$ ), with the value of R ² above 0.9 for both polarization; then, the effects of azimuths and ground characteristics on brightness temperature can also be corrected by establishing their regression relationship with brightness temperature. Combined with these analyses, an empirical model is developed to measure the topographic effect according to RU. The verification results show that the $\Delta TB_{model}$ calculated by the empirical model is similar to the $\Delta TB_{simulated}$ of microwave radiation model (r>0.9), which indicates that the empirical model is suitable for mountainous areas with different terrains. Results also provide a theoretical basis for improving the accuracy of passive microwave remote sensing of mountainous areas.