A TRANSFORMATION METHOD FOR TEXTURE FEATURE DESCRIPTION UNDER DIFFERENT IMAGINE CONDITIONS

Markov random field (MRF) based methods have been widely used in high spatial resolution (HSR) image classification. However, many existing MRF-based methods put more emphasis on pixel level contexts while less on superpixel level contextual information. To cope with this issue, this article presents a novel bilevel contextual MRF framework, named BLC-MRF, for HSR imagery classification. Specifically, pixel and superpixel level dependence are incorporated into the proposed MRF model to fully exploit spectral–spatial contextual information and preserve object boundaries in HSR images. In BLC-MRF, a pixel level MRF model is first performed and then cascaded as an input of a superpixel level MRF. In superpixel level, unary and pairwise potential terms are constructed by using the superpixel probability estimation method and spectral histogram distance, respectively. At last, a contextual MRF model is conducted and the final classification map can be computed by using $\alpha$ -expansion algorithm. The benefits of BLC-MRF are twofold: first, the pixel and superpixel level contextual information can be exploited under MRF framework to preserve object boundaries for improving the classification performance, and, second, the algorithm can provide promising results with a small number of training samples. Experimental results on three HSR datasets demonstrate that the proposed approach outperforms several state-of-the-art methods in terms of the classification performance.

A novel fuzzy decision tree is proposed in this paper (the FDT-support vector machine (SVM) classifier), where the node discriminations are implemented via binary SVMs. The tree structure is determined via a class grouping algorithm, which forms the groups of classes to be separated at each internal node, based on the degree of fuzzy confusion between the classes. In addition, effective feature selection is incorporated within the tree building process, selecting suitable feature subsets required for the node discriminations individually. FDT-SVM exhibits a number of attractive merits such as enhanced classification accuracy, interpretable hierarchy, and low model complexity. Furthermore, it provides hierarchical image segmentation and has reasonably low computational and data storage demands. Our approach is tested on two different tasks: natural forest classification using a QuickBird multispectral image and urban classification using hyperspectral data. Exhaustive experimental investigation demonstrates that FDT-SVM is favorably compared with six existing methods, including traditional multiclass SVMs and SVM-based binary hierarchical trees. Comparative analysis is carried out in terms of testing rates, architecture complexity, and computational times required for the operative phase.

By labelling high spatial resolution (HSR) images with specific semantic classes according to geographical properties, scene classification has been proven to be an effective method for HSR remote sensing image semantic interpretation. Deep learning is widely applied in HSR remote sensing scene classification. Most of the scene classification methods based on deep learning assume that the training datasets and the test datasets come from the same datasets or obey similar feature distributions. However, in practical application scenarios, it is difficult to guarantee this assumption. For new datasets, it is time-consuming and labor-intensive to repeat data annotation and network design. The neural architecture search (NAS) can automate the process of redesigning the baseline network. However, traditional NAS lacks the generalization ability to different settings and tasks. In this paper, a novel neural network search architecture framework—the spatial generalization neural architecture search (SGNAS) framework—is proposed. This model applies the NAS of spatial generalization to cross-domain scene classification of HSR images to bridge the domain gap. The proposed SGNAS can automatically search the architecture suitable for HSR image scene classification and possesses network design principles similar to the manually designed networks. To obtain a simple and low-dimensional search space, the traditional NAS search space was optimized and the human-the-loop method was used. To extend the optimized search space to different tasks, the search space was generalized. The experimental results demonstrate that the network searched by the SGNAS framework with good generalization ability displays its effectiveness for cross-domain scene classification of HSR images, both in accuracy and time efficiency.

Content-based remote sensing (RS) image retrieval (CBRSIR) is a critical way to organize high-resolution RS (HRRS) images in the current big data era. The increasing volume of HRRS images from different satellites and sensors leads to more attention to the cross-source CSRSIR (CS-CBRSIR) problem. Due to the data drift, one crucial problem in CS-CBRSIR is the modality discrepancy. Most existing methods focus on finding a common feature space for various HRRS images to address this issue. In this space, their similarity relations can be measured directly to obtain the cross-source retrieval results straight. This way is feasible and reasonable, however, the specific information corresponding to HRRS images from different sources is always ignored, limiting retrieval performance. To overcome this limitation, we develop a new model for CS-CBRSIR in this paper named dual modality collaborative learning (DMCL). To fully explore the specific information from diverse HRRS images, DMCL first introduces ResNet50 as the feature extractor. Then, a common space mutual learning module is developed to map the specific features into a common space. Here, the modality discrepancy is reduced from the aspects of features and their distributions. Finally, to supplement the specific knowledge to the common features, we develop modality transformation and the dual-modality feature learning modules. Their function is to transmit the specific knowledge from different sources mutually and fuse the specific and common features adaptively. The comprehensive experiments are conducted on a public dataset. Compared with many existing methods, the behavior of our DMCL is stronger. These encouraging results for a public dataset indicate that the proposed DMCL is useful in CS-CBRSIR tasks.

Based on analysis of roads features in high resolution remote sensing (RS) images, we improve the traditional algorithm for automatic edge detection and extraction and put forward a new methodology. Mathematical morphology reconstruction and boundary tracing algorithm are introduced to automatically detect and extract edges from high resolution RS images. Three factors, P factor, F factor and C factor, are defined to describe the shape characteristics of edges. K-mean cluster algorithm is then utilized to identify and extract road edges from the huge edge sets. Finally an example gives fine result of this methodology.

Among the global Earth system changes, human-induced land-cover and land-use changes (LULCC) play significant roles in sustainability of ecosystems. Suburban sprawl, for example, is one of the major components of LULCC that develops urban impervious surface area (ISA) across the landscapes. Increasing ISA is a major environmental issue that has profound impacts on hydrology and climate. Coastal state of Rhode Island has experienced problems associated with urban runoffs due to the alteration of ISA on the landscape. Precise and accurate coverage and location of ISA are key parameters in ISA related environmental studies. Due to the limitations of existing coarse spatial resolution ISA data and current methods for extraction of ISA from high spatial resolution remote sensing imageries, this study developed an algorithm of multiple agent segmentation and classification (MASC) to extract ISA information from high spatial resolution remote sensing imageries. The MASC model includes submodels of segmentation, shadow-effect, MANOV A-based classification, and post-classification. Based on the MASC model, this study built the 1-meter high spatial resolution ISA dataset for the state of Rhode Island, and revealed its spatial patterns. The result indicates that, as of 2004, 10% of the state land has been covered by ISA. The major population centers and historical cities, such as the Providence, Woonsocket, and Newport, have ISA over 30%. The heavily settled suburban communities have ISA between 10% and 30%. Only 17 out of 39 towns in the state have less than 10% ISA. The average ISA for the coastal towns is 14%. Because most stream quality indicators are predicted to decline when watershed ISA exceeds 10%, the results from this study serve as an alarming indicator for managing the state's watershed and coastal ecosystems. It is important to develop scientific basis for observing and modeling watershed hydrologic dynamic with emphasis on how the hydrologic processes are affected by the spatial heterogeneity of ISA, and to improve the understanding of influence of ISA on hydrologic cycle in watersheds. This developed a distributed object-oriented rainfall-runoff simulation (DORS) model that enhances capability and performance of hydrologic modeling with incorporation of high spatial resolution ISA information. With the innovative process of object-oriented spatial units, the DORS model can reduce data volume, increase computational efficiency, strengthen representation of watersheds and utilize the data in variable scales. The DORS model provides a framework to integrate remote sensing data and the derived products in different scale for the simulation of

In this paper, we evaluate the capability of the high spatial resolution airborne Digital Airborne Imaging System (DAIS) imagery for detailed vegetation classification at the alliance level with the aid of ancillary topographic data. Image objects as minimum classification units were generated through the Fractal Net Evolution Approach (FNEA) segmentation using eCognition software. For each object, 52 features were calculated including spectral features, textures, topographic features, and geometric features. After statistically ranking the importance of these features with the classification and regression tree algorithm (CART), the most effective features for classification were used to classify the vegetation. Due to the uneven sample size for each class, we chose a non-parametric (nearest neighbor) classifier. We built a hierarchical classification scheme and selected features for each of the broadest categories to carry out the detailed classification, which significantly improved the accuracy. Pixel-based maximum likelihood classification (MLC) with comparable features was used as a benchmark in evaluating our approach. The objectbased classification approach overcame the problem of saltand-pepper effects found in classification results from traditional pixel-based approaches. The method takes advantage of the rich amount of local spatial information present in the irregularly shaped objects in an image. This classification approach was successfully tested at Point Reyes National Seashore in Northern California to create a comprehensive vegetation inventory. Computer-assisted classification of high spatial resolution remotely sensed imagery has good potential to substitute or augment the present ground-based inventory of National Park lands.

In content based image retrieval (CBIR) process, every image has been represented in a compact set of local visual features i.e. color, texture, and/or shape of images. This set of local visual features is known as feature vector. In the CBIR process, feature vectors of images have been used to represent or to identify similar images in adequate way. As a result, feature vector construction has always been considered as an important issue since it must reflect proper image semantics using minimal amount of data. The proposed CBIR scheme is based on the combination of color and texture features. In this work initially, we have converted the given RGB image into HSV color image. Subsequently, we have considered H (hue), S (saturation), and V (intensity) components for extraction of visual image features. The texture features have been extracted from the V component of the image using 2D dual-tree complex wavelet transform (2D DT-CWT) where it analyzes the textural patterns in six different directions i.e. ±15\(^{\circ }\), ±45\(^{\circ }\), and ±75\(^{\circ }\). At the same time, we have computed the probability histograms of H and S components of the image respectively and subsequently those are divided into non-uniform bins based on cumulative probability for extraction of color based features. So, in this work both the color and texture features have been extracted simultaneously. Finally, the obtained features have been concatenated to attain the final feature vector and same is considered in image retrieval process. We have tested the novelty and performance of the proposed work in two Corel, two objects, and, a texture image datasets. The experimental results reveal the acceptable retrieval performances for different types of datasets.

Greenhouses are densely distributed across the cultivated land in high-resolution remote sensing imagery, resulting in the problem of dense object extraction. On the one hand, objects tend to be wrongly merged into one object since objects are connected; on the other hand, the existed random sampling scheme does not make good use of the object distribution density to improve the training effect. To meet the demand for dense greenhouse extraction, this paper proposes a novel deep learning-based greenhouse extraction algorithm. To solve the problem that dense objects tend to be wrongly merged, this paper proposes a dual-task learning module, which uses the discriminant object boundary of the pixel-based branch to distinguish the objects of the object-based branch; to take advantage of the object distribution density for effective training, a high-density biased sampler is proposed. Moreover, this paper provides a dataset of manually labeled imagery to train and develop the proposed method. Results on greenhouse extraction in six regions in China achieve peak mIoU and mAP value, surpassing the state-of-the-art methods. Finally, a product of a greenhouse map in China is provided for analysis.

In order to improve the separation between abnormal cells and noncellular artifacts in the CERVIFIP automated cervical cytology prescreening system, 22 different object texture features were investigated. The features were all statistical parameters of the pixel density histograms or one-dimensional filtered values of central and border regions of the object images. The features were calculated for 231 images (100 cells and 131 artifacts) detected as Suspect Cells by the current CERVIFIP and were then tested in hierarchical and linear discriminant classifiers. After selecting the two best features for use in a hierarchical classifier, 83% correct classification was achieved. One of these features was specifically designed to remove poorly focused objects. With maximum likelihood discrimination using all 22 features, an overall correct classification rate of 90% was obtained.

Dryland salinity is a major land management issue globally, and results in the abandonment of farmland. Revegetation with halophytic shrub species such as Atriplex nummularia for carbon mitigation may be a viable option but to generate carbon credits ongoing monitoring and verification is required. This study investigated the utility of high-resolution airborne images (Digital Multi Spectral Imagery (DMSI)) obtained in two seasons to estimate carbon stocks at the plant- and stand-scale. Pixel-scale vegetation indices, sub-pixel fractional green vegetation cover for individual plants, and estimates of the fractional coverage of the grazing plants within entire plots, were extracted from the high-resolution images. Carbon stocks were correlated with both canopy coverage (R2: 0.76–0.89) and spectral-based vegetation indices (R2: 0.77–0.89) with or without the use of the near-infrared spectral band. Indices derived from the dry season image showed a stronger correlation with field measurements of carbon than those derived from the green season image. These results show that in semi-arid environments it is better to estimate saltbush biomass with remote sensing data in the dry season to exclude the effect of pasture, even without the refinement provided by a vegetation classification. The approach of using canopy cover to refine estimates of carbon yield has broader application in shrublands and woodlands.

Aiming at the deficiency that blockchain technology is too sensitive to the binary-level changes of high resolution remote sensing (HRRS) images, we propose a new subject-sensitive hashing algorithm specially for HRRS image blockchains. To implement this subject-sensitive hashing algorithm, we designed and implemented a deep neural network model MultiRes-RCF (richer convolutional features) for extracting features from HRRS images. A MultiRes-RCF network is an improved RCF network that borrows the MultiRes mechanism of MultiResU-Net. The subject-sensitive hashing algorithm based on MultiRes-RCF can detect the subtle tampering of HRRS images while maintaining robustness to operations that do not change the content of the HRRS images. Experimental results show that our MultiRes-RCF-based subject-sensitive hashing algorithm has better tamper sensitivity than the existing deep learning models such as RCF, AAU-net, and Attention U-net, meeting the needs of HRRS image blockchains.

Remote sensing change detection (CD) involves identifying differences between two satellite images of the same geographic area taken at different times. It plays a critical role in applications such as urban planning and disaster management. Traditional CD methods rely on manually extracted features, which often lack robustness and accuracy in capturing the details of objects. Recently, deep learning-based methods have expanded the applications of CD in high-resolution remote sensing images, yet they struggle to fully utilize the multi-level features extracted by backbone networks, limiting their performance. To address this challenge, we propose a Semantic-Guided Cross-Attention Network (SCANet). It introduces a Hierarchical Semantic-Guided Fusion (HSF) module, which leverages high-level semantic information to guide low-level spatial details through an attention mechanism. Additionally, we design a Cross-Attention Feature Fusion (CAFF) module to establish global correlations between bitemporal images, thereby improving feature interaction. Extensive experiments on the IWHR-data and LEVIR-CD datasets demonstrate that SCANet significantly outperforms existing State-of-the-Art (SOTA) methods. Specifically, the F1-score and the Intersection over Union (IoU) score are improved by 2.002% and 3.297% on the IWHR-data dataset and by 0.761% and 1.276% on the LEVIR-CD dataset, respectively. These results validate the effectiveness of semantic-guided fusion and cross-attention feature interaction, providing new insights for advancing change detection research in high-resolution remote sensing imagery.

Despite increasing efforts in the mapping of landslides using Sentinel-1 and -2, research on their combination for discerning historical landslides in forest areas is still lacking, particularly using object-oriented machine learning approaches. This study was accomplished to test the efficiency of Sentinel-derived features and digital elevation model (DEM) derivatives for mapping old and new landslides, using object-oriented random forest. Two forest subsets were selected including a protected and non-protected forest in northeast Iran. Landslide samples were obtained from CORONA images and aerial photos (old landslides), and also field mensuration and high-resolution images (new landslides). Segment objects were generated from a set combination of Sentinel-1A, Sentinel-2A, and some topographic-derived indices using multiresolution segmentation algorithm. Various object features were derived from the main channels of Sentinel images and DEM derivatives in the seven main groups, including spectral layers, spectral indices, geometric, contextual, textural, topographic, and hydrologic features. A single database was created, including landslide samples and Sentinel- and DEM-derived object features. Roughly 20% of landslide-affected objects and non-landslide-affected objects were randomly selected as an input for training the random forest classifier. Two-thirds of the selected objects were assigned as learning samples for classification, and the remainder were used for testing the accuracy of landslide and non-landslide classification. Results indicated that: (1) The sensitivity of mapping historical landslides was 86.6% and 80.3% in the protected and non-protected forests, respectively; (2) the object features of Sentinel-2A and DEM obtained the highest importance with the total scores of 55.6% and 32%, respectively in the protected forests, and 65.4% and 21% respectively in the non-protected forests; (3) the features derived from the combination of Sentinel-1 and -2A demonstrated a total importance of 10% for mapping new landslides; and (4) textural features were obtained in approximately two-thirds of the total scores for mapping new landslides, however a combination of topographic, spectral, textural, and contextual features were the effective predictors for mapping old landslides. This research proposes applying a synergetic analysis of Sentinel- and DEM-derived features for mapping historical landslides; however, there are no uniformly pre-defined influential variables for mapping historical landslides in different forest areas.

Road network extraction plays a critical role in data updating, urban development, and decision support. To improve the efficiency of labeling road datasets and addressing the problems of traditional methods of manually extracting road networks from high-resolution images, such as their slow speed and heavy workload, this paper proposes a semi-automatic method of road network extraction from high-resolution remote-sensing images. The proposed method needs only a few points to extract a single road in the image. After the roads are extracted one by one, the road network is generated according to the width of each road and the spatial relationships among the roads. For this purpose, we use regional growth, morphology, vector tracking, vector simplification, endpoint modification, road connections, and intersection connections to generate road networks. Experiments on four images with different terrains and different resolutions show that this method has high extraction accuracy under different image conditions. The comparisons with the semi-automatic GVF-snake method based on regional growth also showed its advantages and potentiality. The proposed method is a novel form of semi-automatic road network extraction, and it significantly increases the efficiency of road network extraction.