An Object-Based Method for Road Network Extraction in VHR Satellite Images

This paper presents a method for road extraction from lidar data based on support vector machine (SVM) classification. The lidar data are used exclusively to evaluate the potential in the road extraction process. First, the SVM algorithm is used to classify the lidar data into five classes: road, tree, building, grassland, and cement. Then, some misclassified pixels in the road class is removed using the road values in the normalized Digital Surface Model and Normalized Difference Distance features. In the postprocessing stage, a method based on Radon transform and Spline interpolation is employed to automatically locate and fill the gaps in the road network. The experimental results show that the proposed algorithm for gap filling works well on straight roads. The proposed road extraction algorithm is tested on three datasets. An accuracy assessment indicated 63.7%, 60.26% and 66.71% quality for three datasets. Finally, centerline of the detected roads is extracted using mathematical morphology. Road information plays an important role in many modern applications, including transportation, automatic navigation systems, traffic management, and crisis management, and enables existing geographic information system (GIS) databases to be updated more efficiently. In the past two decades, automatic road extraction has become an important topic in remote sensing, photogrammetry, and computer vision. In addition, recent advances in lidar systems and their enormous potential in automatic feature extraction motivate the development of automatic road extraction algorithms based on lidar data.

As the acquisition of very high resolution (VHR) satellite images becomes easier owing to technological advancements, ever more stringent requirements are being imposed on automatic image interpretation. Moreover, per-pixel classification has become the focus of research interests in this regard. However, the efficient and effective processing and the interpretation of VHR satellite images remain a critical task. Convolutional neural networks (CNNs) have recently been applied to VHR satellite images with considerable success. However, the prevalent CNN models accept input data of fixed sizes and train the classifier using features extracted directly from the convolutional stages or the fully connected layers, which cannot yield pixel-to-pixel classifications. Moreover, training a CNN model requires large amounts of labeled reference data. These are challenging to obtain because per-pixel labeled VHR satellite images are not open access. In this paper, we propose a framework called the unsupervised-restricted deconvolutional neural network (URDNN). It can solve these problems by learning an end-to-end and pixel-to-pixel classification and handling a VHR classification using a fully convolutional network and a small number of labeled pixels. In URDNN, supervised learning is always under the restriction of unsupervised learning, which serves to constrain and aid supervised training in learning more generalized and abstract feature. To some degree, it will try to reduce the problems of overfitting and undertraining, which arise from the scarcity of labeled training data, and to gain better classification results using fewer training samples. It improves the generality of the classification model. We tested the proposed URDNN on images from the Geoeye and Quickbird sensors and obtained satisfactory results with the highest overall accuracy (OA) achieved as 0.977 and 0.989, respectively. Experiments showed that the combined effects of additional kernels and stages may have produced better results, and two-stage URDNN consistently produced a more stable result. We compared URDNN with four methods and found that with a small ratio of selected labeled data items, it yielded the highest and most stable results, whereas the accuracy values of the other methods quickly decreased. For some categories with fewer training pixels, accuracy for categories from other methods was considerably worse than that in URDNN, with the largest difference reaching almost 10%. Hence, the proposed URDNN can successfully handle the VHR image classification using a small number of labeled pixels. Furthermore, it is more effective than state-of-the-art methods.

This research investigates the use of scale-space theory to detect individual trees in orchards from very-high resolution (VHR) satellite images. Trees are characterized by blobs, for example, bell-shaped surfaces. Their modeling requires the identification of local maxima in Gaussian scale space, whereas location of the maxima in the scale direction provides information about the tree size. A two-step procedure relates the detected blobs to tree objects in the field. First, a Gaussian blob model identifies tree crowns in Gaussian scale space. Second, an improved tree crown model modifies this model in the scale direction. The procedures are tested on the following three representative cases: an area with vitellaria trees in Mali, an orchard with walnut trees in Iran, and one case with oil palm trees in Indonesia. The results show that the refined Gaussian blob model improves upon the traditional Gaussian blob model by effectively discriminating between false and correct detections and accurately identifying size and position of trees. A comparison with existing methods shows an improvement of 10–20% in true positive detections. We conclude that the presented two-step modeling procedure of tree crowns using Gaussian scale space is useful to automatically detect individual trees from VHR satellite images for at least three representative cases.

Superpixels, as a state-of-the-art segmentation paradigm, have recently been widely used in computer vision and pattern recognition. Despite the effectiveness of these algorithms, there are still many limitations and challenges dealing with Very High-Resolution (VHR) satellite images especially in complex urban scenes. In this paper, we develop a superpixel algorithm as a modified edge-based version of Simple Linear Iterative Clustering (SLIC), which is here called ESLIC, compatible with VHR satellite images. Then, based on the modified properties of generated superpixels, a heuristic multi-scale approach for building extraction is proposed, based on the stereo satellite imagery along with the corresponding Digital Surface Model (DSM). First, to generate the modified superpixels, an edge-preserving term is applied to retain the main building boundaries and edges. The resulting superpixels are then used to initially refine the stereo-extracted DSM. After shadow and vegetation removal, a rough building mask is obtained from the normalized DSM, which highlights the appropriate regions in the image, to be used as the input of a multi-scale superpixel segmentation of the proper areas to determine the superpixels inside the building. Finally, these building superpixels with different scales are integrated and the output is a unified building mask. We have tested our methods on building samples from a WorldView-2 dataset. The results are promising, and the experiments show that superpixels generated with the proposed ESLIC algorithm are more adherent to the building boundaries, and the resulting building mask retains urban object shape better than those generated with the original SLIC algorithm.

Satellite imagery has been used to improve scientific research worldwide. In this study, the southern right whale (Eubalaena australis) was chosen to test the use of medium, high, and very high resolution (VHR) satellite images, on the Brazilian breeding ground. These images were used to identify the whales and were compared to aerial survey data collected in the same area. The VHR satellite images from the Pleiades‐1A satellite, available on Google Earth, displayed the best results when compared to those from Sentinel 2, Landsat 8, Rapid Eye, and Planet Scope. No significant differences were observed (Mann‐Whitney U test) between the possible whales recorded in the satellite images and of real whales recorded in situ by aerial surveys, while considering either the number of groups (p = .841, n = 5) or the total number of animals (p = .222, n = 5). Further, when using VHR images, the geographical positions of the whales recorded in situ had a positive correlation with the positions generated from the satellite images (Mantel test: r = 0.52, p = .001, n = 13). This technique may represent an important tool for detecting right whales, especially in countries where research funding is scarce.

Historical maps are valuable sources of geospatial data for various geography-related applications, providing insightful information about historical land use, transportation infrastructure, and settlements. While transformer-based segmentation methods have been widely applied to image segmentation tasks, they have mostly focused on satellite images. There is a growing need to explore transformer-based approaches for geospatial object extraction from historical maps, given their superior performance over traditional convolutional neural network (CNN)-based architectures. In this research, we aim to automatically extract five different road types from historical maps, using a road dataset digitized from the scanned Deutsche Heereskarte 1:200,000 Türkei (DHK 200 Turkey) maps. We applied the variants of the transformer-based SegFormer model and evaluated the effects of different encoders, batch sizes, loss functions, optimizers, and augmentation techniques on road extraction performance. Our best results, with an intersection over union (IoU) of 0.5411 and an F1 score of 0.7017, were achieved using the SegFormer-B2 model, the Adam optimizer, and the focal loss function. All SegFormer-based experiments outperformed previously reported CNN-based segmentation models on the same dataset. In general, increasing the batch size and using larger SegFormer variants (from B0 to B2) resulted in improved accuracy metrics. Additionally, the choice of augmentation techniques significantly influenced the outcomes. Our results demonstrate that SegFormer models substantially enhance true positive predictions and resulted in higher precision metric values. These findings suggest that the output weights could be directly applied to transfer learning for similar historical maps and the inference of additional DHK maps, while offering a promising architecture for future road extraction studies.

Case instance segmentation of small farmland based on Mask R-CNN of feature pyramid network with double attention mechanism in high resolution satellite images

In this article, we develop a novel method for the detection of vineyard parcels in agricultural landscapes based on very high resolution (VHR) optical remote sensing images. Our objective is to perform texture-based image retrieval and supervised classification algorithms. To do that, the local textural and structural features inside each image are taken into account to measure its similarity to other images. In fact, VHR images usually involve a variety of local textures and structures that may verify a weak stationarity hypothesis. Hence, an approach only based on characteristic points, not on all pixels of the image, is supposed to be relevant. This work proposes to construct the local extrema-based descriptor (LED) by using the local maximum and local minimum pixels extracted from the image. The LED descriptor is formed based on the radiometric, geometric and gradient features from these local extrema. We first exploit the proposed LED descriptor for the retrieval task to evaluate its performance on texture discrimination. Then, it is embedded into a supervised classification framework to detect vine parcels using VHR satellite images. Experiments performed on VHR panchromatic PLEIADES image data prove the effectiveness of the proposed strategy. Compared to state-of-the-art methods, an enhancement of about 7% in retrieval rate is achieved. For the detection task, about 90% of vineyards are correctly detected.

With geospatial intelligence research developing rapidly, automatic road extraction is becoming a fundamental and challenging task. Due to the special geometric structure and spectral information of road networks, existing methods suffer from incomplete and fractured results. In this work, a novel road extraction convolutional neural network, incorporating the road boundary details and road junction information via a dual-branch multi-task structure, is proposed to learn synergistic feature representations and strengthen road connectivity. Firstly, a BiFPN-based feature aggregation module is utilised to bridge the semantic gap between low-level and high-level feature maps, allowing multi-scale spatial details to be fully fused. Secondly, the boundary auxiliary branch, using a U-shaped network with a spatial-channel attention module, captures residential information for the backbone to enhance the subtleties of road edges. Thirdly, the node inferring branch models the road junction position jointly with the road surface, aiming to strengthen the topology structure and reduce the fragmented road segments. We perform experiments on three diverse road datasets, namely the DeepGlobe dataset, Massachusetts dataset, and SpaceNet dataset. The results demonstrate that our model shows an overall performance improvement over some SOTA algorithms and the IoU indicator achieves 3.86%, 0.79%, and 1.71% improvements over Unet on the three datasets, respectively. • A novel Geometric Prior Assisted Neural Network (GA-Net) is proposed for automatic road extraction. The boundary information is introduced to the model, which strengthens the geometric details of road edges. The road junction information is jointly learned with the road surface information, which reduces the missing gap between road segments. • A feature fusion module is designed to bridge the semantic gap between multi-scale features maps. An attention module is designed to strengthen the road geometric information. The two light modules can be easily ensemble with other models. • We achieves better results than some SOTA algorithms for three popular datasets. And results of our model shows better completeness and connectivity.

This paper deals with automatic road extraction from airborne SAR imagery. Automatic extracted road networks are often incomplete. Even sophisticated road extraction approaches that are developed for optical imagery often fail to extract a complete road network. In SAR imagery in contrary to optical imagery, roads are more affected by high backscattering context objects. These local context objects like trees, bridges, or vehicles can disturb road extraction, but they can also support it. We show that an explicit modeling of context objects, leads to improved extraction results.

Abstract. The road extraction in urban areas is difficult task due to the complicated patterns and many contextual objects. LiDAR data directly provides three dimensional (3D) points with less occlusions and smaller shadows. The elevation information and surface roughness are distinguishing features to separate roads. However, LiDAR data has some disadvantages are not beneficial to object extraction, such as the irregular distribution of point clouds and lack of clear edges of roads. For these problems, this paper proposes an automatic road centerlines extraction method which has three major steps: (1) road center point detection based on multiple feature spatial clustering for separating road points from ground points, (2) local principal component analysis with least squares fitting for extracting the primitives of road centerlines, and (3) hierarchical grouping for connecting primitives into complete roads network. Compared with MTH (consist of Mean shift algorithm, Tensor voting, and Hough transform) proposed in our previous article, this method greatly reduced the computational cost. To evaluate the proposed method, the Vaihingen data set, a benchmark testing data provided by ISPRS for “Urban Classification and 3D Building Reconstruction” project, was selected. The experimental results show that our method achieve the same performance by less time in road extraction using LiDAR data.

The number of high and very high resolution (VHR) optical satellite sensors, as well as the number of medium resolution satellites is continuously growing. However, not all high-resolution optical satellite imaging cameras have a sufficient and stable calibration in time. Due to their high agility in rotation, a quick change in viewing direction can lead to satellite attitude oscillation, causing image distortions and thus affecting image geometry and geo-positioning accuracy. This paper presents an approach based on re-projection of regularly distributed 3D ground points from object in image space, to detect and estimate the periodic distortions of Pléiades tri-stereo imagery caused by satellite attitude oscillations. For this, a hilly region was selected as a test site. Consequently, we describe a complete processing pipeline for computing the systematic height errors (deformations, waves) of the satellite-based digital elevation model by using a Lidar high resolution terrain model. Ground points with fixed positions, but with two elevations (actual and corrected) are then re-projected to the satellite images with the aid of the Rational Polynomial Coefficients (RPCs) provided with the imagery. Therefore, image corrections (displacements) are determined by computing the differences between the distinct positions of corresponding points in image space. Our experimental results in Allentsteig (Lower Austria) show that the systematic height errors of satellite-based elevation models cannot be compensated with an usual or even high number of Ground Control Points (GCPs) for RPC bias correction, due to insufficiently known image orientations. In comparison to a reference Lidar Digital Terrain Model (DTM), the computed elevation models show undulation effects with a maximum height difference of 0.88 m in along-track direction. With the proposed method, image distortions in-track direction with amplitudes of less than 0.15 pixels were detected. After applying the periodic distortion compensation to all three images, the systematic elevation discrepancies from the derived elevation models were successfully removed and the overall accuracy in open areas improved by 33% in the RMSE. Additionally, we show that a coarser resolution reference elevation model (AW3D30) is not feasible for improving the geometry of the Pléiades tri-stereo satellite imagery.

An automatic method for road extraction in rural and semi-urban areas starting from high resolution satellite imagery

Oil palm trees detection and counting on Microsoft Bing Maps Very High Resolution (VHR) satellite imagery and Unmanned Aerial Vehicles (UAV) data using image processing thresholding approach

Reliable building delineation in very high resolution (VHR) satellite imagery can be achieved by precise disparity information extracted from stereo pairs. However, off-nadir VHR images over urban areas contain many occlusions due to building leaning that creates gaps in the extracted disparity maps. The typical approach to fill these gaps is by interpolation. However, it inevitably degrades the quality of the disparity map and reduces the accuracy of building detection. Thus, this research proposes a registration-based technique for mapping the disparity of off-terrain objects to avoid the need for disparity interpolation and normalization. The generated disparity by the proposed technique is then used to support building detection in off-nadir VHR satellite images. Experiments in a high-rise building area confirmed that 75 percent of the detected building roofs overlap precisely the reference data, with almost 100 percent correct detection. These accuracies are substantially higher than those achieved by other published research.