Geographic Object-Based Image Analysis (GEOBIA) techniques have become increasingly popular in remote sensing. GEOBIA has been claimed to represent a paradigm shift in remote sensing interpretation. Still, GEOBIA—similar to other emerging paradigms—lacks formal expressions and objective modelling structures and in particular semantic classification methods using ontologies. This study has put forward an object-based semantic classification method for high resolution satellite imagery using an ontology that aims to fully exploit the advantages of ontology to GEOBIA. A three-step workflow has been introduced: ontology modelling, initial classification based on a data-driven machine learning method, and semantic classification based on knowledge-driven semantic rules. The classification part is based on data-driven machine learning, segmentation, feature selection, sample collection and an initial classification. Then, image objects are re-classified based on the ontological model whereby the semantic relations are expressed in the formal languages OWL and SWRL. The results show that the method with ontology—as compared to the decision tree classification without using the ontology—yielded minor statistical improvements in terms of accuracy for this particular image. However, this framework enhances existing GEOBIA methodologies: ontologies express and organize the whole structure of GEOBIA and allow establishing relations, particularly spatially explicit relations between objects as well as multi-scale/hierarchical relations.

Remote sensing (RS) image segmentation is an essential step in geographic object-based image analysis (GEOBIA) to ultimately derive “meaningful objects”. While many segmentation methods exist, most of them are not efficient for large data sets. Thus, the goal of this research is to develop an efficient parallel multi-scale segmentation method for RS imagery by combining graph theory and the fractal net evolution approach (FNEA). Specifically, a minimum spanning tree (MST) algorithm in graph theory is proposed to be combined with a minimum heterogeneity rule (MHR) algorithm that is used in FNEA. The MST algorithm is used for the initial segmentation while the MHR algorithm is used for object merging. An efficient implementation of the segmentation strategy is presented using data partition and the “reverse searching-forward processing” chain based on message passing interface (MPI) parallel technology. Segmentation results of the proposed method using images from multiple sensors (airborne, SPECIM AISA EAGLE II, WorldView-2, RADARSAT-2) and different selected landscapes (residential/industrial, residential/agriculture) covering four test sites indicated its efficiency in accuracy and speed. We conclude that the proposed method is applicable and efficient for the segmentation of a variety of RS imagery (airborne optical, satellite optical, SAR, high-spectral), while the accuracy is comparable with that of the FNEA method.

Comprehensively understanding the factors influencing crime is a prerequisite for preventing and combating crime. Although some studies have investigated the relationship between environmental factors and property crime, the interaction between factors was not fully considered in these studies, and the explanation of complex factors may be insufficient. This paper explored the influence of environmental factors on property crime using factor regression and factor interaction based on data from the central city of Lanzhou, China. Our findings showed that: (1) The distribution of crime cases showed the pattern of a local multi-center. Shop density, hotel density, entertainment density and house price were the four dominant environmental drivers of property crime; (2) The relationship between the light intensity and property crime had different correlation explanations in temporal projection and spatial projection. There was a normal distribution curve between the number of property crimes and the Price-to-Earnings Ratio (PE Ratio) of the community house price; and (3) The results of the factor interaction indicated that the effect of all factors on crime showed a two-factor enhancement. As an important catalyst, shop density had the strongest interaction with other factors. Shop density gradient influenced the degree of interpretation of spatial heterogeneity of property crime.

Hyperspectral image classification (HSIC) is widely used in such fields as vegetation classification and fine agriculture. Nowadays, numerous classification models based on Transformer networks are inputting data typically in the form of patches. However, such patch-based input format may weaken the spatial relationships between neighbor pixels, thus limiting the ability to obtain contextual semantic information. Additionally, hyperspectral images (HSIs) generally contain hundreds of bands, which in turn include some redundant information. The original spectral vectors may lead to redundancy of information, thus reducing the classification accuracy remarkably. Therefore, in this paper, a Patch-based Transformer Network Construction with Adaptive Feature-interaction for HSIC called AFinet is developed for HSIC. As an end-to-end network, AFinet consists of a feature extraction module and an adaptive feature-interaction (AFi) module. For the developed model, the feed-forward neural network (FNN) in the standard Transformer framework suffers from a limited ability in exploiting local contexts. In this paper, an expression-enhanced feed-forward neural network (E2FNN) is introduced, which can incorporate depthwise convolution layers to capture local contextual information, so as to enhance the correlations between neighbor pixels. Moreover, this study designs an AFi module to facilitate the sharing of interaction features among Transformer-extracted features. Subsequently, global spatial feature information can be integrated into each spectral channel. The AFi module also includes a channel attention mechanism to help focus more attention on channels that contain critical information, while paying less attention to channels with little critical information. After the approach proposed in this study was tested on three widely used datasets, the experimental results demonstrate that the AFinet developed in this study can effectively improve classification accuracy.

High-resolution remote sensing change detection methods typically rely on single-modal image features, making them highly susceptible to pseudo-changes induced by factors like illumination, shadows, and seasonal variations. Moreover, most approaches adopt a unified binary change paradigm, overlooking the heterogeneous mechanisms of change, such as appearance and disappearance. To address these limitations, this paper presents KGDNet, a knowledge-guided tri-branch network designed for decoupled modeling of heterogeneous change mechanisms in change detection (CD). KGDNet leverages annotation-free monocular depth maps as geometric priors and, through the Guided Alignment and Fusion Module (GAFM), upgrades 2D features to 3D geometry-aware representations. This enhances the discrimination of genuine land-cover changes while effectively suppressing pseudo-changes at the source. The network explicitly decouples three core mechanisms—appearance, difference, and disappearance—using specialized parallel branches to capture distinct change patterns. Additionally, the difference branch dynamically corrects and refines prior knowledge, mitigating contamination from low-quality depth cues. To efficiently process high-resolution imagery, we further propose a Multi-Head Latent Spatiotemporal Attention (MH-LSTA) module that compresses and remaps features in a latent space, substantially reducing computational and memory costs while enhancing the modeling of global spatiotemporal relations. This yields a favorable trade-off between accuracy and efficiency. Comprehensive experiments on public benchmarks demonstrate that KGDNet outperforms representative state-of-the-art (SOTA) methods in key metrics, delivering superior detection accuracy and robustness. This study proposes a knowledge-guided decoupled modeling paradigm that integrates prior knowledge with heterogeneous change mechanisms to enhance the interpretability of change representations. The code and data will be released upon publication at https://github.com/yangzy1217/KGDNet.