Extracting farmland from high spatial resolution remote sensing images is a basic task for agricultural information management. According to Tobler’s first law of geography, closer objects have a stronger relation. Meanwhile, due to the scale effect, there are differences on both spatial and attribute scales among different kinds of objects. Thus, it is not appropriate to segment images with unique or fixed parameters for different kinds of objects. In view of this, this paper presents a stratified object-based farmland extraction method, which includes two key processes: one is image region division on a rough scale and the other is scale parameter pre-estimation within local regions. Firstly, the image in RGB color space is converted into HSV color space, and then the texture features of the hue layer are calculated using the grey level co-occurrence matrix method. Thus, the whole image can be divided into different regions based on the texture features, such as the mean and homogeneity. Secondly, within local regions, the optimal spatial scale segmentation parameter was pre-estimated by average local variance and its first-order and second-order rate of change. The optimal attribute scale segmentation parameter can be estimated based on the histogram of local variance. Through stratified regionalization and local segmentation parameters estimation, fine farmland segmentation can be achieved. GF-2 and Quickbird images were used in this paper, and mean-shift and multi-resolution segmentation algorithms were applied as examples to verify the validity of the proposed method. The experimental results have shown that the stratified processing method can release under-segmentation and over-segmentation phenomena to a certain extent, which ultimately benefits the accurate farmland information extraction.

Landslide susceptibility mapping (LSM) is an effective way to predict spatial probability of landslide occurrence. Existing convolutional neural network (CNN)-based methods apply self-built CNN with simple structure, which failed to reach CNN's full potential on high-level feature extraction, meanwhile ignored the use of numerical predisposing factors. For the purpose of exploring feature fusion based CNN models with greater reliability in LSM, this study proposes an ensemble model based on channel-expanded pre-trained CNN and traditional machine learning model (CPCNN-ML). In CPCNN-ML, pre-trained CNN with mature structure is modified to excavate high-level features of multichannel predisposing factor layers. LSM result is generated by traditional machine learning (ML) model based on hybrid feature of high-level features and numerical predisposing factors. Lantau Island, Hong Kong is selected as study area; temporal landslide inventory is used for model training and evaluation. Experimental results show that CPCNN-ML has ability to predict landslide occurrence with high reliability, especially the CPCNN-ML based on random forest. Contrast experiments with self-built CNN and traditional ML models further embody the superiority of CPCNN-ML. It is worth noting that coastal regions are newly identified landslide-prone regions compared with previous research. This finding is of great reference value for Hong Kong authorities to formulate appropriate disaster prevention and mitigation policies.

Automatic landslide detection based on very high spatial resolution (VHR) remote sensing images is crucial for disaster prevention and mitigation applications. With the rapid development of deep learning techniques, state-of-the-art semantic segmentation methods based on fully convolutional network (FCNN) have achieved outstanding performance in the landslide detection task. However, most of the existing studies only utilize visual features. Even if the advanced FCNN models are applied, there is still a certain amount of falsely detected and miss detected landslides. In this article, we innovatively introduce landslide susceptibility as prior knowledge and propose an innovative susceptibility-guided landslide detection method based on FCNN (SG-FCNN) to detect landslides from single temporal images. In addition, an unsupervised change detection method based on the mean changing magnitude of objects (MCMO) is further proposed and integrated with the SG-FCNN to detect newly occurred landslides from bi-temporal images. The effectiveness of the proposed SG-FCNN and MCMO has been tested in Lantau Island, Hong Kong. The experimental results show that the SG-FCNN can significantly reduce the amount of falsely detected and miss detected landslides compared with the FCNN. It can conclude that applying landslide susceptibility as prior knowledge is much more effective than using visual features only, which introduces a new methodology of landslide detection and lifts the detection performance to a new level.

In recent years, torrential rainfall has triggered numerous shallow landslides in southeastern China. Therefore, giving a spatial-temporal hazard prediction for rainfall-induced shallow landslides on a fine-grained scale is imperative. Nowadays, most empirical hazard prediction methods are qualitative and consider only rainfall features. The generated prediction results lack sufficient spatial details and have a high false alarm rate. By improving the traditional hazard prediction method based on frequency statistics and utilizing landslide susceptibility, this paper proposes a quantitative hazard prediction method that simultaneously considers landslide triggering and predisposing factors. In the case of predicting natural terrain landslide hazards in Hong Kong, the spatial accuracy and fineness of the prediction results generated by the proposed methods are significantly enhanced compared to those predicted by the qualitative method based on rainfall features, thereby confirming the efficacy, reliability, and benefits of the proposed methods.

?й?ˮ????ѧ??????ͨ???ڿ????߱༭ϵͳ??????Ͷ??