In the problem of ship detection in complex scenes, in addition to the characteristics of ship targets, there is rich semantic information in the global and local background of the whole scenes, which provides more valuable inference information for ship detection. Therefore, in this paper, we propose a ship detection method in complex scenes considering both global and local information perception for SAR images. Firstly, the proposed method detects the globally stable region and the locally significant region respectively, and then designs a judgment method combining the two to eliminate false alarms, so as to ensure that the detected target has both globally stable characteristics and locally significant characteristics. The detection performance of the proposed method is verified by the spaceborne SAR images covering the coastal areas. The result shows that the proposed method can effectively detect ships in complex scenes, especially eliminating most false alarms in land areas.

Recently, multistatic airborne SAR (MuA-SAR) is becoming a research hotspot due to its flexibility. Multi-platform data fusion requires that the imaging algorithm has strong adaptability to the flight path and relative spatial configuration of the airborne platforms. Therefore, the time domain algorithm based on back projection (BP) is suitable. However, in the existing BP-based methods, data needs to be projected into each grid one by one. In fact, not all pixels are target pixels that need to be projected, and the back projection of non-target pixels leads to a lot of invalid computation. Applying these methods directly to MuA-SAR will inevitably lead to a great increase in computation. To reduce the redundant back projection operation of BP algorithm and improve the efficiency of imaging processing in MuA-SAR, a fast imaging method based on segmentation before imaging is proposed in this paper. On the basis of fast factorized back projection (FFBP) algorithm architecture, an image segmentation method based on maximally stable extremal regions (MSER) is introduced. In the process of recursive fusion at each stage, only the pixel information of the segmented suspected target area is transferred to the next stage for fusion, and then the imaging efficiency is improved. The simulation and comparative experiments verify the effectiveness of the proposed method.

Multistatic airborne synthetic aperture radar (MuA-SAR) systems can achieve high-resolution imaging in a short time by fusing observation data from multiple radar platforms. However, its imaging quality relies on a rigorous design of the spatial configuration (SC) of each platform, mainly including the relative spatial separation and velocity. The rigorously designed SCs make it difficult to obtain in actual flight and weaken the flexibility advantage brought by the airborne platforms. Therefore, it is meaningful and necessary to explore a new SC design method to obtain relaxed SCs under the condition of ensuring imaging quality. In this paper, to relax the limitations of SC, an optimal design method for MuA-SAR SC is proposed. First, the relationship between the spatial configuration, wavenumber spectrum (WS) distribution, and imaging performance is established, and it visually reveals the configuration limitations. Second, an optimized search space of SC is defined by the peak to sidelobe ratio (PSLR) to relax the space to compromised configurations. Finally, the SC design problem is transformed into a constrained multiple objective optimization problem (CMOP) which is solved by the multiple objective particle swarm optimization (MOPSO) algorithm. The simulation results show that the proposed method can still obtain the optimized SC beyond the strictly restricted configuration space, which expands the SC limitations of the MuA-SAR system.

In the field of remote sensing, harbor detection in SAR images has an important application prospect. However, the complex coastline of SAR images increases the difficulty of harbor detection. In response to this problem, a cascaded harbor detection (CHD) method for SAR image based on corner and coastline features is proposed in this paper. First, coast-line is extracted from SAR image by sea-land segmentation. Then, in the first step rough detection, corner detection is performed on the coastline and the detected corners are automatically clustered to locate the harbor candidate areas. Finally, the second step precise detection is carried out on the coast-line of harbor candidate areas, where coastline feature detection is completed by using corners again to remove the fake harbor targets in harbor candidate areas. Experimental results based on satellite-borne SAR data prove the proposed CHD method enjoys a preferable detection performance compared with existing harbor detection methods.

Ship target segmentation plays an important role in synthetic aperture radar (SAR) image interpretation. However, existing segmentation methods for marine SAR images have the problem of inaccurate edge segmentation, a concern for real-world applications. In this letter, we propose a clustering center shifted adaptive target segmentation (CCSATS) method. Firstly, the proposed clustering center shift method is used to update the clustering centers of each iteration, which can quickly and accurately capture ship pixels. Then, based on regional homogeneity coefficients, we define a new similarity measurement criterion with two adaptive weight factors to ensure the homogeneity of segmentation results. Finally, neighborhood patches are used to represent pixel information, which can reduce the influence of speckle noise and enhance the target edge fitting ability. Our segmentation results of measured SAR images show that the proposed method effectively ensures segmentation accuracy. Compared with other existing methods, the proposed target segmentation method achieves better edge capture performance.

SAR image segmentation is a key step of SAR image interpretation, boosting target detection and recognition. Since similar targets may exist in complex and changeable scenes, under-segmentation and over-segmentation often occur in SAR image segmentation. To solve the above deficiencies, we propose a superpixel aggregation method based on multi-direction gray level co-occurrence matrix (GLCM) for SAR image segmentation. Firstly, a linear similarity judgment based on gray feature and spatial distance of pixels is introduced. In this stage, we expand the search range of clustering centers and add constraints to reduce the deviation, so as to alleviate over-segmentation. Then, for the spatial adjacent su-perpixels, we use multi-direction GLCM to measure texture similarity between them, merging homogeneous superpixel-s to solve under-segmentation. Experimental results based on satellite-borne SAR images from different scenes illustrate that the proposed method performs well with excellent pixel accuracy, effectively solving under-segmentation and over-segmentation.

In SAR image target detection, harbor detection can help the detection of harbor targets and maritime traffic planning. In this paper, we propose a harbor detection method of SAR images based on multidirectional one-dimensional scanning. Take the candidate points along the coastline and the multidirectional one-dimensional scanning is performed. Using the distribution characteristics of land, sea and dock in the one-dimensional vector, training a convolutional neural network to classify the candidate points into harbor and non-harbor feature points. Then we get the harbor feature points map reflecting the distribution of harbors. The Sentinel-1 spaceborne SAR images covering a coastal region are used to verify the proposed method. The experimental results show the effectiveness and accuracy of the proposed method.

Synthetic Aperture Radar (SAR) is widely used in environmental monitoring and disaster early warning due to its high resolution imaging performance. A distributed radar system can be established by mounting radars on multiple unmanned aerial vehicle (UAV) platforms. Distributed radar utilizes multiple transmitters distributed in different spatial positions, flying along a certain planned path and enable multiple transmitters to obtain as large an aperture as possible in a certain time. In this paper, an intelligent optimal path planning method for distributed radar short-time aperture synthesis is proposed, which can deal with terrain obstacles and line-of-sight occlusion in UAV flight path and achieve the goal of maximum aperture accumulation in a specific time. Simulation results verified the effectiveness of the UAV intelligent optimal path planning method.

This paper proposes a parking space information monitoring system by millimeter wave synthetic aperture radar (SAR) based on unmanned aerial vehicle (UAV). Parking space information that people are concerned about includes vacant parking place, parking place occupied by obstacles and place parked by vehicles. Specially, the free parking space detection is an important module for the parking guidance system (PGS) that can help drivers to find parking space efficiently. In this system, we obtain high resolution SAR images of parking lots at first. Then, in order to define the free parking space, Maximally Stable Extremal Region (MSER) method is exploited to leach the candidate regions occupied by vehicles from millimeter wave SAR images. Next, the system utilize visual saliency detection method to extract obstacles from the non-parked parking space acquired by pre-detection. Ultimately, the three types of information have been determined, including vacant parking space, parking space occupied by obstacles and the parked place. Experimental results prove that the integrated scheme performs well in parking information determination.

Finding a suitable parking position often leads to much traffic pressure and time consumption in a busy parking lot. An auxiliary parking method based on automotive millimeter wave SAR is proposed in this paper. Firstly, Maximally Stable Extremal Region (MSER) method is utilized to extract the candidate regions occupied by parked vehicles from the millimeter wave SAR images. Then, in order to eliminate the false alarm candidate regions, we employ the morphological filter and utilize the centroid position to further refine the candidate regions. Thirdly, the difference in width-to-height ratio of the candidate regions is exploited to distinguish the parking directions of the cars. After that, the available parking spaces are located according to the parking direction. Finally, further remove the spaces occupied by obstacles, and plan reasonable parking routes. Experimental results based on measured data show that the proposed method has outstanding detection and parking route planning performance in different scenes.

SAR ship detection is essential for marine monitoring. Due to the high similarity between the harbor and the ship body on gray and texture features, the traditional methods are unable to achieve effective inshore ship detection. An improved Faster R-CNN based on MSER decision criterion for SAR ship detection in harbor is proposed in this paper. It is a ship detection method based on the combination of feature-based method and pixel-based method. Firstly, Faster R-CNN is used to generate region proposals. Then, replace the threshold decision criterion of Faster R-CNN with the maximum stability extremal region (MSER) method to reassess the generated region proposals with higher scores, aiming at improving the detection rate and reducing the false alarm rate simultaneously. Experimental results based on satellite-borne SAR data illustrate that the proposed method obtains excellent detection performance and low false alarm rate.

Distributed radar imaging utilizes expanded array elements in space to form a large aperture and obtain high imaging resolution. A great number of array elements are required in traditional distributed radar system which uses multiple platforms. The distribution of spatial spectrum is affected by the number and the signal form of array elements. In this research, to improve the utilization efficiency of platform resources, a resource allocation optimization method based on Unmanned Aerial Vehicle(UAV) is proposed. It chooses the optimized bandwidth and sampling frequency points of array elements by analyzing the relationship between spatial spectrum and imaging performance. This method can use a small number of UAVs to maintain high imaging resolution. Simulation results verified the effectiveness of the resource allocation optimization method for image quality improvement.