Sea-land Segmentation Research Articles

Sea-Crossing Bridge Detection in Polarimetric SAR Images Based on Windowed Level Set Segmentation and Polarization Parameter Discrimination

As sea-crossing bridges are important hubs connecting separated land areas, their detection in SAR images is of great significance. However, under complex scenarios, the sea surface conditions, the distribution of coastal terrain morphologies, and the scattering components of different structures in the bridge area are very complex and diverse, which makes the accurate and robust detection of sea-crossing bridges difficult, including the sea–land segmentation and bridge feature extraction on which the detection depends. In this paper, we propose a polarimetric SAR image detection method for sea-crossing bridges based on windowed level set segmentation and polarization parameter discrimination. Firstly, the sea and land are segmented by a proposed windowed level set segmentation method, which replaces the construction of the level set segmentation energy function based on the isolated pixel distribution with a joint distribution of pixels in a certain window region. Secondly, water regions of interest are extracted by a proposed water region merging algorithm combining the distances of the water contour and polarization similarity parameter. Finally, the bridge regions of interest (ROIs) are extracted by merging close water contours, and the ROIs are discriminated by the polarimetric parameters of the polarization entropy and scattering angle. Experimental results using multiple AirSAR, RADARSAT-2, and TerraSAR-X quad-polarization SAR data from the coastal areas of San Francisco in the USA, Singapore, and Fuzhou, Fujian, and Zhanjiang, Guangdong, in China show that the proposed method can achieve 100% detection of sea-crossing bridges in different bands for different scenes, and the accuracy of the intersection of the ground-truth (IoG) index of bridge body recognition can reach more than 85%. The proposed method can improve the detection rate and reduce the false alarm rate compared with the traditional spatial-based method.

Sea–Land Segmentation Using HED-UNET for Monitoring Kaohsiung Port

In recent years, it has become a trend to analyze shoreline changes through satellite images in coastal engineering research. The results of sea–land segmentation are very important for shoreline detection. CoastSat is a time-series shoreline detection system that uses an artificial neural network (ANN) on sea–land segmentation. However, the method of CoastSat only uses the spectral features of a single pixel and ignores the local relationships of adjacent pixels. This impedes optimal category prediction, particularly considering interference by climate features such as clouds, shadows, and waves. It is easy to cause the classifier to be disturbed in the classification results, resulting in classification errors. To solve the problem of misclassification of sea–land segmentation caused by climate interference, this paper applies HED-UNet to the image dataset obtained from CoastSat and learns the relationship between adjacent pixels by training the deep network architecture, thereby improving the results of erroneous sea–land segmentation due to climate disturbances. By using different optimizers and loss functions in the HED-Unet model, the experiment verifies that Adam + Focal loss has the best performance. The results also show that the deep learning model, HED-Unet, can effectively improve the accuracy of the sea–land segmentation to 97% in a situation with interference from atmospheric factors such as clouds and waves.

Gaussian Function Fusing Fully Convolutional Network and Region Proposal-Based Network for Ship Target Detection in SAR Images

Recently, ship target detection in Synthetic aperture radar (SAR) images has become one of the current research hotspots and plays an important role in the real-time detection of sea regions. The traditional SAR ship detection methods usually consist of two modules, one module named land-sea segmentation for removing the complicated land regions, and one module named ship target detection for realizing fine ship detection. An algorithm combining the Unet-based land-sea segmentation method and improved Faster RCNN-based ship detection method is proposed in this paper. The residual convolution module is introduced into the Unet structure to deepen the network level and improve the feature representation ability. The K-means method is introduced in the Faster RCNN method to cluster the size and aspect ratio of ship targets, to improve the anchor frame design, and make it more suitable for our ship detection task. Meanwhile, a fine detection algorithm uses the Gaussian function to fuse the confidence value of sea-land segmentation results and the coarse detection results. The segmentation and detection results on the established segmentation dataset and detection dataset, respectively, demonstrate the effectiveness of our proposed segmentation methods and detection methods.

Fast and accurate sea-land segmentation of SAR imagery based on multi-source data registration

Targeted on accurate and efficient ship detection of Synthetic Aperture Radar imagery, sea-land segmentation is key for the complex port scenes. Despite the importance of this task, current methods are still severely influenced by several limitations, such as high level of image speckle noise, complicated backscattering characteristics. In order to overcome these issues, we propose a sea-land segmentation method of Synthetic Aperture Radar imagery based on multi-source data registration. To be specific, high-resolution optical remote sensing images and high-precision coastline vector data are integrated with Synthetic Aperture Radar images for accurate sea-land segmentation. First, the Pyramid Level-by-level Guided-Histogram of the Orientated Phase Congruency method is developed for the registration of Synthetic Aperture Radar and optical images, making two types of images into the same spatial reference. Following this, the high-precision coastline vector data collected from optical images is transferred to the Synthetic Aperture Radar image, realizing accurate sea-land segmentation. Finally, aiming at ship detection tasks, two quantitative evaluation indicators for sea-land segmentation are proposed: (1) the Absolute Segmentation Accuracy; and (2) the Ship Mis-segmentation Index. The accuracy and efficiency of the proposed method are tested and verified on Gaofen-3 SAR images of Yokosuka Port, Japan. Experimental results show that the proposed method achieves an Absolute Segmentation Accuracy of 3.15 pixels, a Ship Mis-segmentation Index of 9.23% and processing time of 82.92 s on average, which reveals that the proposed method outperforms current popular approaches in terms of accuracy and efficiency.

Automatic coastline extraction through enhanced sea-land segmentation by modifying Standard U-Net

Sea-land segmentation (SLS) is an essential remote sensing task for various coastal and environmental studies such as coastline extraction, coastal erosion, coastal area monitoring, and ship or iceberg detection. This study aims at improving the SLS performance by modifying the Standard U-Net (SUN) model and developing an automatic coastline extraction framework. SUN generally has an acceptable performance in many applications. However, better SLS outputs are needed for reliable coastline extraction. In our proposed framework, we firstly analyzed three different input images, including Red-Green-Blue (RGB), Normalized Difference Water Index (NDWI), and Near-Infrared (NIR) images. Secondly, we modified the SUN architecture to improve the segmentation results. The main modifications are using different loss functions and two fusion methods for RGB and NIR images. The segmentation results were then passed into the subsequent automatic coastline extraction pipeline based on morphological operations and pixel connectivity analysis. The training and testing steps were accomplished utilizing a benchmark dataset of China’s coastal areas. Moreover, another dataset consisting of a time series of Landsat-8 imagery from the southern Caspian Sea coastlines was collected to evaluate coastline extraction efficiency. The results indicate that the proposed modifications could effectively enhance the performance of the SUN, with the most significant improvement to the Intersection over Union (IoU) score being as high as 1.68% and 8.95% in China and Caspian Sea datasets, respectively, while outperforming other state-of-the-art models including FC-DenseNet and DeepLabV3+.

Small Vessel Detection Based on Adaptive Dual-Polarimetric Feature Fusion and Sea–Land Segmentation in SAR Images

Detection of small sea vessels in synthetic aperture radar (SAR) images has received much attention in recent years because the small vessels have weak scattering intensity and few image pixels. The existing detection network structures are not well adapted to small-scale targets, the polarimetric data are not properly utilized, and the sea–land segmentation process to remove land false alarms is time-consuming. Regarding these problems, first, a single low-level path aggregation network is designed specifically for small targets. The structure reduces false alarms at the feature level by finding suitable single-scale feature maps for detection and adding a semantic enhancement module. Second, adaptive dual-polarimetric feature fusion is proposed to filter the multichannel features acquired by dual-polarimetric decomposition to reduce feature redundancy. Third, a segmentation layer is added to the network to shield the land from false alarms. The detection and segmentation layers share the feature extraction and feature fusion modules and are jointly trained by a joint loss. Finally, polarimetric SAR detection and segmentation dataset containing small vessel detection and sea–land segmentation labels is created with reference to the LS-SSDDv1.0 dataset, and experimental results on this dataset verify the improvement of this proposed method over other typical methods.

Multi-Scale Sea-Land Segmentation Method for Remote Sensing Images Based on Res2Net

Multi-Scale Sea-Land Segmentation Method for Remote Sensing Images Based on Res2Net

BS-Net: Using Joint-Learning Boundary and Segmentation Network for Coastline Extraction from Remote Sensing Images

ABSTRACT The coastline extraction from remote-sensing images is of great significance to the dynamic monitoring of the coastal zone. The types of coastlines are complex and diverse, and they show different spectrum, texture, and shape features, so accurately extracting coastlines is still a challenging task. The semantic segmentation model based on deep learning has good generalization ability. However, the down sampling operation will lose the location of boundary information, resulting in the location offset between the extracted coastlines and the actual coastlines. A multi-task network, called the joint learning network of boundary and segmentation (BS-Net), was proposed in this letter. BS-Net adds a coastline positioning stream to supervise the location of the coastlines. Moreover, this letter designed a boundary-segmentation interaction (BSI) module for the mutual guidance of information between the coastline positioning stream and the sea-land segmentation stream to correct the coastline features and enhance the segmentation boundary. The experimental results on a set of Gaofen-1 remote sensing images showed that, for various natural coastlines and artificial coastlines, coastlines extracted based on BS-Net were more accurate than those extracted by other methods. Code is available at: https://github.com/weiAI1996/BS-Net.

Ship Detection and Feature Visualization Analysis Based on Lightweight CNN in VH and VV Polarization Images

Synthetic aperture radar (SAR) is a significant application in maritime monitoring, which can provide SAR data throughout the day and in all weather conditions. With the development of artificial intelligence and big data technologies, the data-driven convolutional neural network (CNN) has become widely used in ship detection. However, the accuracy, feature visualization, and analysis of ship detection need to be improved further, when the CNN method is used. In this letter, we propose a two-stage ship detection for land-contained sea area without a traditional sea-land segmentation process. First, to decrease the possibly existing false alarms from the island, an island filter is used as the first step, and then threshold segmentation is used to quickly perform candidate detection. Second, a two-layer lightweight CNN model-based classifier is built to separate false alarms from the ship object. Finally, we discuss the CNN interpretation and visualize in detail when the ship is predicted in vertical–horizontal (VH) and vertical–vertical (VV) polarization. Experiments demonstrate that the proposed method can reach an accuracy of 99.4% and an F1 score of 0.99 based on the Sentinel-1 images for a ship with a size of less than 32 × 32.

A Novel Coarse-to-Fine Sea-Land Segmentation Technique Based on Superpixel Fuzzy C-Means Clustering and Modified Chan-Vese Model

The sea-land segmentation for optical remote sensing images (RSIs) has a valuable role in water resources and coastal zones management. However, it is challenging because optical RSIs mainly suffer from low contrast, intensity inhomogeneity with mixed pixels, and large image size with redundant information. This paper introduces a novel coarse-to-fine sea-land segmentation method that incorporates Superpixel Fuzzy C-Means (SPFCM) and a Modified Chan-Vese active contour model (MCV). First, the image is over-segmented into superpixels to reduce the information redundancy and utilize spectral and spatial information. The SPFCM employs local relationships among neighboring superpixels to cluster the superpixels based on their color and texture features, deal with mixed pixels, and produce coarse segmentation results. The CV depends on the initial contour. If the contour is incorrectly initialized, the CV may trap in local minima and needs many iterations. The purpose of the SPFCM result is to provide an automatic initial contour for an MCV model instead of manual initialization to improve the CV model's performance. Finally, color and texture features are combined in vector-valued images to solve the traditional CV's problems to deal with complicated nature images with intensity inhomogeneity or rich texture features to produce fine segmentation results. The proposed method is evaluated and achieved an average accuracy of 98.9%, an average Jaccard Similarity Coefficient (JSC) equals 97.1%, an average Disc Similarity Coefficient (DSC) equals 98.5%, and an average recall equals 99.3%. The proposed method results are promising, and it outperformed the results of other state-of-the-art sea-land segmentation methods.

SANet: A Sea–Land Segmentation Network Via Adaptive Multiscale Feature Learning

Sea-land segmentation of remote sensing images is of great significance to the dynamic monitoring of coastlines. However, the types of objects in the coastal zone are complex, and their spectra, textures, shapes, and distribution features are different. Therefore, sea-land segmentation for various types of coastlines is still a challenging task. In this article, a scale-adaptive semantic segmentation network, called SANet, is proposed for sea-land segmentation of remote sensing images. SANet has made two innovations on the basis of the classic encoder-decoder structure. First, to integrate the spectral, textural, and semantic features of ground objects at different scales, we designed an adaptive multiscale feature learning module (AML) to replace the conventional serial convolution operation. The AML module mainly contains a multiscale feature extraction unit and an adaptive feature fusion unit. The former can capture the multiscale detailed information and contextual semantic information of objects from an early stage, while the latter can adaptively fuse feature maps of different scales. Second, we adopted the squeeze-and-excitation module to bridge the corresponding layers of the codec so that SANet can selectively emphasize the features of the weak sea-land boundaries. Experiments on a set of Gaofen-1 remote sensing images demonstrated that SANet achieved more accurate segmentation results and obtained sharper boundaries than other methods for various natural and artificial coastlines.

A Novel Coarse-to-Fine Method of Ship Detection in Optical Remote Sensing Images Based on a Deep Residual Dense Network

Automatic ship detection in optical remote sensing images is of great significance due to its broad applications in maritime security and fishery control. Most ship detection algorithms utilize a single-band image to design low-level and hand-crafted features, which are easily influenced by interference like clouds and strong waves and not robust for large-scale variation of ships. In this paper, we propose a novel coarse-to-fine ship detection method based on discrete wavelet transform (DWT) and a deep residual dense network (DRDN) to address these problems. First, multi-spectral images are adopted for sea-land segmentation, and an enhanced DWT is employed to quickly extract ship candidate regions with missing alarms as low as possible. Second, panchromatic images with clear spatial details are used for ship classification. Specifically, we propose the local residual dense block (LRDB) to fully extract semantic feature via local residual connection and densely connected convolutional layers. DRDN mainly consists of four LRDBs and is designed to further remove false alarms. Furthermore, we exploit the multiclass classification strategy, which can overcome the large intra-class difference of targets and identify ships of different sizes. Extensive experiments demonstrate that the proposed method has high robustness in complex image backgrounds and achieves higher detection accuracy than other state-of-the-art methods.

Phase spectrum based automatic ship detection in synthetic aperture radar images

This paper proposes an automatic ship detection approach in Synthetic Aperture Radar (SAR) Images using phase spectrum. The proposed method mainly contains two stages: Firstly, sea-land segmentation of SAR Images is one of the key stages for SAR image application such as sea-targets detection and recognition, which are easily detected only in sea regions. In order to eliminate the influence of land regions in SAR images, a novel land removing method is explored. The removing method employs a Harris corner detector to obtain some image patches belonging to land, and the probability density function (PDF) of land area can be estimated by these patches. Thus, an appropriate land segmentation threshold is accordingly obtained. Secondly, an automatic ship detector based on phase spectrum is proposed. The proposed detector is free from various idealized assumptions and can accurately detect ships in SAR images. Experimental results demonstrate the efficiency of the proposed ship detection algorithm in diversified SAR images.

High-Throughput, Resource-Efficient Multi-Dimensional Parallel Architecture for Space-Borne Sea-Land Segmentation

Sea-land segmentation based on edge detection is commonly utilized in ship detection, coastline extraction, and satellite system applications due to its high accuracy and rapid speed. Pixel-level distribution statistics do not currently satisfy the requirements for high-resolution, large-scale remote sensing image processing. To address the above problem, in this paper, we propose a high-throughput hardware architecture for sea-land segmentation based on multi-dimensional parallel characteristics. The proposed architecture is well suited to wide remote sensing images. Efficient multi-dimensional block level statistics allow for relatively infrequent pixel-level memory access; a boundary block tracking process replaces the whole-image scanning process, markedly enhancing efficiency. The tracking efficiency is further improved by a convenient two-step scanning strategy that feeds back the path state in a timely manner for a large number of blocks in the same direction appearing in the algorithm. The proposed architecture was deployed on Xilinx Virtex k7-410t to find that its practical processing time for a [Formula: see text] remote sensing image is only about 0.4[Formula: see text]s. The peak performance is 1.625[Formula: see text]gbps, which is higher than other FPGA implementations of segmentation algorithms. The proposed structure is highly competitive in processing wide remote sensing images.

Sea–Land Segmentation in Maritime Surveillance Radars via K-Nearest Neighbor Classifier

Shipborne or airborne maritime surveillance radars at scan mode work at a complex scene consisting of land, sea, and islands. Sea–land segmentation provides a two-class classification of sea clutter versus ground clutter to the surveillance scene as a precondition of adaptive target detection. It is a difficult problem because only a few coherent pulses are available at scan mode. Moreover, due to moving radar platforms and wide dynamic range of ground and sea clutter in power, average amplitude, Doppler offset, and initial phase of radar, the returns vector fails to distinguish sea clutter and ground clutter. In this article, a similarity measure of two radar returns vectors, which is invariant to amplitude, Doppler offset, and initial phase, is constructed, which is closely relevant to the Doppler bandwidth of a returns vector. Based on the similarity measure, a K-nearest neighbor classifier is proposed to yield a pixel-level sea–land segmentation of the scene. Further, the morphological filtering is operated on the pixel-level segmentation to obtain a region-level segmentation. Moreover, the discrete Frechet distances of the main boundaries in successive scan periods are used to assess segmentation quality. The proposed method is verified by measured data from an airborne radar and an island-based radar. The results show that it behaves better than the methods using thresholding phase linearity degree of radar returns and the support vector machine, and back propagation neural network in a three-dimensional feature space.

An Improved Descalloping Method Combined With Imaging Parameters for GaoFen-3 ScanSAR

Due to the specific working mechanism of alternately transmitting and receiving signals between multiple adjacent swaths, scanning synthetic aperture radar (ScanSAR) will cause periodical wavelike modulation of the intensity image along azimuth direction, which is known as scalloping. Conventional descalloping methods are achieved by proper azimuth antenna pattern (AAP) correction and multi-looking techniques but are limited by the accuracy of Doppler centroid estimating and the thermal noise. Another type of method extracts and suppresses the scalloping texture on the image, but the scanning parameters of ScanSAR are insufficiently considered. The period of scalloping on the image is related to the period of switching between subswathes. While the harmonics can be calculated by the period of a periodic signal, an improved frequency filtering method combined with imaging parameters is proposed. The scalloping modulation model of ScanSAR combined with imaging parameters is constructed, and the harmonics of scalloping texture are accurately calculated and filtered. For the low signal to noise ratio (SNR) image, the antenna pattern is modified according to SNR to avoid scalloping caused by noise. For non-uniform scenes, scalloping suppression is achieved by using the scalloping features acquired by uniform scenes. To separate the non-uniform scenes from uniform scenes, we still use the characteristics of harmonics caused by scalloping. Our descalloping method achieves accurate suppression of scalloping without sea-land segmentation and ship mask and the residual scalloping is reduced from 1.0 to 0.3–0.5 dB. The residual scalloping and statistical characteristics of the image are analyzed to demonstrate the effectiveness of the proposed method.

Sea-Land Segmentation Using Deep Learning Techniques for Landsat-8 OLI Imagery

Automated coastline extraction from optical satellites is fundamental to coastal mapping, and sea-land segmentation is the core technology of coastline extraction. Deep convolutional neural networks (DCNNs) have performed well in semantic segmentation in recent years. However, sea-land segmentation using deep learning techniques remains a challenging task, due to the lack of a benchmark dataset and the difficulty of deciding which semantic segmentation model to use. We present a comparative framework of sea-land segmentation to Landsat-8 OLI imagery via semantic segmentation in deep learning techniques. Three issues are investigated: (1) constructing a sea-land benchmark dataset using Landsat-8 Operational Land Imager (OLI) imagery consisting of 18,000 km2 of coastline around China; (2) evaluating the feasibility and performance of sea-land segmentation by comparing the accuracy assessment, time complexity, spatial complexity and stability of state-of-the-art DCNNs methods; (3) choosing the most suitable semantic segmentation model for sea-land segmentation in accordance with Akaike information criterion (AIC) and Bayesian information criterion (BIC) model selection. Results show that the average test accuracy achieves over 99% accuracy, and the mean Intersection over Unions (mean IoU) is above 92%. These findings demonstrate that the Fully Convolutional DenseNet (FC-DenseNet) performs better than other state-of-the-art methods in sea-land segmentation, based on both AIC and BIC. Considering training time efficiency, DeeplabV3+ performs better for sea-land segmentation. The sea-land segmentation benchmark dataset is available at: https://pan.baidu.com/s/1BlnHiltOLbLKe4TG8lZ5xg.

Current Trends and Future Directions in Sea-Land Segmentation for Remote Sensing Images

Current Trends and Future Directions in Sea-Land Segmentation for Remote Sensing Images

A Novel Deep Structure U-Net for Sea-Land Segmentation in Remote Sensing Images

Sea-land segmentation is an important process for many key applications in remote sensing. Proper operative sea-land segmentation for remote sensing images remains a challenging issue due to complex and diverse transition between sea and lands. Although several Convolutional Neural Networks (CNNs) have been developed for sea-land segmentation, the performance of these CNNs is far from the expected target. This paper presents a novel deep neural network structure for pixel-wise sea-land segmentation, a Residual Dense U-Net (RDU-Net), in complex and high-density remote sensing images. RDU-Net is a combination of both down-sampling and up-sampling paths to achieve satisfactory results. In each down- and up-sampling path, in addition to the convolution layers, several densely connected residual network blocks are proposed to systematically aggregate multi-scale contextual information. Each dense network block contains multilevel convolution layers, short-range connections and an identity mapping connection which facilitates features re-use in the network and makes full use of the hierarchical features from the original images. These proposed blocks have a certain number of connections that are designed with shorter distance backpropagation between the layers and can significantly improve segmentation results whilst minimizing computational costs. We have performed extensive experiments on two real datasets Google Earth and ISPRS and compare the proposed RDUNet against several variations of Dense Networks. The experimental results show that RDUNet outperforms the other state-of-the-art approaches on the sea-land segmentation tasks.

Squeeze and Excitation Rank Faster R-CNN for Ship Detection in SAR Images

Synthetic aperture radar (SAR) ship detection is an important part of marine monitoring. With the development in computer vision, deep learning has been used for ship detection in SAR images such as the faster region-based convolutional neural network (R-CNN), single-shot multibox detector, and densely connected network. In SAR ship detection field, deep learning has much better detection performance than traditional methods on nearshore areas. This is because traditional methods need sea–land segmentation before detection, and inaccurate sea–land mask decreases its detection performance. Though current deep learning SAR ship detection methods still have many false detections in land areas, and some ships are missed in sea areas. In this letter, a new network architecture based on the faster R-CNN is proposed to further improve the detection performance by using squeeze and excitation mechanism. In order to improve performance, first, the feature maps are extracted and concatenated to obtain multiscale feature maps with ImageNet pretrained VGG network. After region of interest pooling, an encoding scale vector which has values between 0 and 1 is generated from subfeature maps. The scale vector is ranked, and only top $K$ values will be preserved. Other values will be set to 0. Then, the subfeature maps are recalibrated by this scale vector. The redundant subfeature maps will be suppressed by this operation, and the detection performance of detector can be improved. The experimental results based on Sentinel-1 images show that the detection performance of the proposed method achieves 0.836 which is 9.7% better than the state-of-the-art method when using F1 as matric and executes 14% faster.