Synthetic Aperture Radar Datasets Research Articles

The Parallel SBAS Approach for Sentinel-1 Interferometric Wide Swath Deformation Time-Series Generation: Algorithm Description and Products Quality Assessment

We present an advanced differential synthetic aperture radar (SAR) interferometry (DInSAR) processing chain, based on the Parallel Small BAseline Subset (P-SBAS) technique, for the efficient generation of deformation time series from Sentinel-1 (S-1) interferometric wide (IW) swath SAR data sets. We first discuss an effective solution for the generation of high-quality interferograms, which properly accounts for the peculiarities of the terrain observation with progressive scans (TOPS) acquisition mode used to collect S-1 IW SAR data. These data characteristics are also properly accounted within the developed processing chain, taking full advantage from the burst partitioning. Indeed, such data structure represents a key element in the proposed P-SBAS implementation of the S-1 IW processing chain, whose migration into a cloud computing (CC) environment is also envisaged. An extensive experimental analysis, which allows us to assess the quality of the obtained interferometric products, is presented. To do this, we apply the developed S-1 IW P-SBAS processing chain to the overall archive acquired from descending orbits during the March 2015-April 2017 time span over the whole Italian territory, consisting in 2740 S-1 slices. In particular, the quality of the final results is assessed through a large-scale comparison with the GPS measurements relevant to nearly 500 stations. The mean standard deviation value of the differences between the DInSAR and the GPS time series (projected in the radar line of sight) is less than 0.5 cm, thus confirming the effectiveness of the implemented solution. Finally, a discussion about the performance achieved by migrating the developed processing chain within the Amazon Web Services CC environment is addressed, highlighting that a two-year data set relevant to a standard S-1 IW slice can be reliably processed in about 30 h.The presented results demonstrate the capability of the implemented P-SBAS approach to efficiently and effectively process large S-1 IW data sets relevant to extended portions of the earth surface, paving the way to the systematic generation of advanced DInSAR products to monitor ground displacements at a very wide spatial scale.

AN ASSESSMENT OF LAND COVER CHANGE DYNAMICS OF GAJA CYCLONE IN COASTAL TAMIL NADU, INDIA USING SENTINEL 1 SAR DATASET

Abstract. Land cover change is a dynamic phenomenon addressing environmental issues including natural calamities. Recent advancements in geospatial technology and availability of remote sensor data have fostered monitoring and mapping of land cover changes more precisely. Remote sensing is widely used where emerging research findings are focused mainly on coastal hazard studies. Tropical cyclones being an extreme weather event are more powerful and hazardous to southern parts of the Indian subcontinent. Aftermath of the cyclone is extreme causing land cover changes like defoliation, water logging, destruction of cultivable lands, plantations shrub vegetation, dissolving salt pans etc. The tropical cyclones are fierce to devastate the coastal districts of Tamil Nadu and make it a prey to these cyclones. In this paper, an attempt has been made to assess the pre and post cyclonic land cover change by utilizing potential microwave Synthetic Aperture Radar (SAR) dataset. The study portrays the occurrence of a severe cyclonic storm named Gaja that was formed over Bay of Bengal which hit Tamil Nadu on 15th of November 2018 causing high death toll and demolition. The study focuses on the pre and post damage assessment provoked by Gaja cyclone. For analysis, a methodical procedure was followed by utilizing the Sentinel 1 SAR dataset. Random Forest (RF) classifier approach was incorporated for mapping land cover types as it reduces the variance among the classes thus yielding accurate predictions. Results demonstrate that classified imagery using dual polarization SAR dataset outperforms well for RF classifier thus escalating the overall accuracy.

The ASI Integrated Sounder-SAR System Operating in the UHF-VHF Bands: First Results of the 2018 Helicopter-Borne Morocco Desert Campaign

This work is aimed at showing the present capabilities and future potentialities of an imaging radar system that can be mounted onboard flexible aerial platforms, such as helicopters or small airplanes, and may operate in the UHF and VHF frequency bands as Sounder and as Synthetic Aperture Radar (SAR). More specifically, the Sounder operates at 165 MHz, whereas the SAR may operate either at 450 MHz or at 860 MHz. In the work, we present the first results relevant to a set of Sounder and SAR data collected by the radar during a helicopter-borne campaign conducted in 2018 over a desert area in Erfoud, Morocco, just after the conclusion of a system upgrading procedure. In particular, a first analysis of the focusing capabilities of the Sounder mode and of the polarimetric and interferometric capabilities of the SAR mode is conducted. The overall system, originally developed by CO.RI.S.T.A. according to a ASI funding set up in 2010, has been upgraded in the frame of a contract signed in 2015 between ASI and different private and public Italian Research Institutes and Universities, namely CO.RI.S.T.A., IREA-CNR, Politecnico di Milano and University of Trento.

A Novel Vessel Velocity Estimation Method Using Dual-Platform TerraSAR-X and TanDEM-X Full Polarimetric SAR Data in Pursuit Monostatic Mode

In this paper, we demonstrate that the spaceborne dual-platform TerraSAR-X (TSX) and TanDEM-X (TDX) pursuit monostatic mode full polarimetric (full-pol) synthetic aperture radar (SAR) data with a time lag can be used to monitor maritime traffic. For single polarization (single-pol) SAR data, the performance of vessel velocity estimation is mainly determined by 2-D cross correlation of SAR intensity data. As the sea clutter is changing dynamically during the TSX/TDX data acquisition, the correlation between two dual-platform images decreases significantly. We may get unstable or incorrect estimations of vessel velocity, especially under a higher wind condition. For solving this problem, we propose an object-oriented polarimetric likelihood ratio test (PolLRT) method based on the complex Wishart distribution. The proposed method makes PolLRT statistics of the detected target pixels for eliminating the effect of varied sea clutter. Two pairs of full-pol SAR data sets covering the Strait of Gibraltar acquired by dual-platform TSX/TDX in pursuit monostatic mode with a time lag of approximately 10 s are selected for the experiments. The experimental results demonstrate that the proposed PolLRT method has a better performance than that of the classical normalized cross correlation (NCC) method with VV polarization SAR data and the mutual information (MI) method with full-pol SAR data. Specifically, under the lower wind condition, the correct estimation rate of the NCC, the MI, and the proposed PolLRT methods are 85.7%, 57.1%, and 100%, respectively; under the relatively higher wind condition, the correct estimation rate of the above three methods are 48.8%, 23.2%, and 90.1%, respectively.

EVALUATION OF THE PERFORMANCE OF SAR AND SAR-OPTICAL FUSED DATASET FOR CROP DISCRIMINATION

Abstract. Crop discrimination and acreage play a vital role in interpreting the cropping pattern, statistics of the produce and market value of each product. Sultan Battery is an area where a large amount of irrigated and rainfed paddy crops are grown along with Rubber, Arecanut and Coconut. In addition, the northern region of Sultan Battery is covered with evergreen and deciduous forest. In this study, the main objective is to evaluate the performance of optical and Synthetic Aperture Radar (SAR)-optical hybrid fusion imageries for crop discrimination in Sultan Bathery Taluk of Wayanad district in Kerala. Seven land use classes such as paddy, rubber, coconut, deciduous forest, evergreen forest, water bodies and others land use (e.g., built-up, barren etc.) were selected based on literature review and local land use classification policy. Both Sentinel-2A (optical) and sentinel-1A (SAR) satellite imageries of 2017 for Kharif season were used for classification using three machine learning classifiers such as Support Vector Machine (SVM), Random Forest (RF) and Classification and Regression Trees (CART). Further, the performance of these techniques was also compared in order to select the best classifier. In addition, spectral indices and textural matrices (NDVI, GLCM) were extracted from the image and best features were selected using the sequential feature selection approach. Thus, 10-fold cross-validation was employed for parameter tuning of such classifiers to select best hyperparameters to improve the classification accuracy. Finally, best features, best hyperparameters were used for final classification and accuracy assessment. The results show that SVM outperforms the RF and CART and similarly, Optical+SAR datasets outperforms the optical and SAR satellite imageries. This study is very supportive for the earth observation scientists to support promising guideline to the agricultural scientist, policy-makers and local government for sustainable agriculture practice.

Stacked Fisher autoencoder for SAR change detection

Stacked autoencoder is effective in image denoising and classification when it is used for synthetic aperture radar (SAR) change detection. However, the resulting features may not be discriminative enough for in some sense. To alleviate this problem, in this paper we propose a stacked Fisher autoencoder (SFAE) for SAR change detection. Specifically, in the framework of SFAE, unsupervised layer-wise feature learning and supervised fine-tuning are jointly performed when training the network. The trained network can be used to detect the changes in both of the single and multi-polarization SAR datasets in real-time. The proposed SFAE has two advantages. The first one is to expand the stacked autoencoder to suit the environment with the multiplicative noise in SAR change detection. The second is that the features extracted by SFAE are more discriminative than the original stacked autoencoder due to that Fisher discriminant criterion is incorporated into SFAE. The results on the simulated and real SAR datasets indicate that the proposed SFAE algorithm has a significant advantage on multitemporal single/multi-polarization SAR (SAR/PolSAR) change detection.

Multitemporal SAR Image Despeckling Based on a Scattering Covariance Matrix of Image Patch.

This paper presents a despeckling method for multitemporal images acquired by synthetic aperture radar (SAR) sensors. The proposed method uses a scattering covariance matrix of each image patch as the basic processing unit, which can exploit both the amplitude information of each pixel and the phase difference between any two pixels in a patch. The proposed filtering framework consists of four main steps: (1) a prefiltering result of each image is obtained by a nonlocal weighted average using only the information of the corresponding time phase; (2) an adaptively temporal linear filter is employed to further suppress the speckle; (3) the final output of each patch is obtained by a guided filter using both the original speckled data and the filtering result of step 3; and (4) an aggregation step is used to tackle the multiple estimations problem for each pixel. The despeckling experiments conducted on both simulated and real multitemporal SAR datasets reveal the pleasing performance of the proposed method in both suppressing speckle and retaining details, when compared with both advanced single-temporal and multitemporal SAR despeckling techniques.

Research on Loess Landslide Identification, Monitoring and Failure Mode with InSAR Technique in Heifangtai, Gansu

The interferometric synthetic aperture radar (InSAR) technique is used over Heifangtai loess terrace, Gansu province of China to map the distribution of potential loess landslides, the evolution of landslide deformation and the failure mode. Firstly, the archived synthetic aperture radar (SAR) datasets with different spatial resolutions and wavelengths from December 2006 to November 2017 are used to identify the potential landslides. Tens potential landslide areas are identified from December 2006 to March 2011 and from January 2016 to November 2016. Field investigation and optical remote sensing images validate the reliability and accuracy of the identified landslides. Then, the TerraSAR-X data with high spatial and temporal resolution are used to monitor the time series deformation of the typical unstable slopes. Results demonstrate that the landslides with the large accumulative deformation all occur in the following time, and the acceleration dates of failed landslides are successfully captured by InSAR time series results. Finally, two-dimensional deformation monitoring of loess landslide is conducted by combining with ascending and descending SAR datasets. The landslide failure mode are analyzed in depth according to the obtained two-dimensional deformation results, topographic map and remote sensing images. The accuracy of the obtained result is verified by field investigation.

Superpixel cosegmentation algorithm for synthetic aperture radar image change detection

Image segmentation has gained increasing popularity and has been widely applied in various image processing tasks recently. A simple linear iterative clustering (SLIC) superpixel generating algorithm for optical images based on k-means clustering approach works well in terms of computational simplicity and segmentation speed. However, due to the inherent speckle noise, it may generate poor segmentations for synthetic aperture radar (SAR) images. In this paper, an improved similarity measure combining pixel intensity and location similarity with edge information is utilised to replace the Euclidean distance in CIELAB colour space for performing local clustering to generate superpixels by using SLIC. Additionally, a constructed image is used to perform the superpixel cosegmentation, simultaneously segmenting the SAR image pairs. The generated superpixels can be taken as basic units for the subsequent change detection. The experimental results conducted on one simulated dataset and one real-world SAR dataset demonstrate the feasibility and effectiveness of the proposed algorithm.

Monitoring urban land surface deformation (2004–2010) from InSAR, groundwater and levelling data: A case study of Changzhou city, China

Land deformation caused by the overexploitation of groundwater has become a potential geological hazard in Changzhou, China. To reduce the potential damage from deformation disasters, it is quite necessary to monitor them. In this study, C-band ENVISAT ASAR synthetic aperture radar (SAR) datasets acquired from 2004 to 2010 were processed with the small baseline subset (SBAS) interferometric SAR (InSAR) method to investigate the spatial–temporal distribution of land deformation. Levelling survey data were used to assess the deformation accuracy measured by the InSAR method, and groundwater level data were used to analyse the incentives for deformation. The SBAS InSAR results found several subsidence regions in the Wujin district and the central part of the Zhonglou district. The annual deformation rate calculated from the C-band interferograms ranged from −65 to +35 mm/yr in the line of sight (LOS) direction, and the cumulative subsidence ranged from −0.35 to 0.13 m in the vertical direction. The levelling survey data show that the two measurements are generally consistent. Correlation analysis between the groundwater table and InSAR measurements at six groundwater well stations shows that groundwater recharges can cause land rebound within the Changzhou urban region. However, subsidence was also observed simultaneously with a rising water table, which might have been caused by the status of consolidation of the strata. Our study provides scientific evidence on the management of groundwater extraction and the assessment of land-subsidence hazard.

Dual and Single Polarized SAR Image Classification Using Compact Convolutional Neural Networks

Accurate land use/land cover classification of synthetic aperture radar (SAR) images plays an important role in environmental, economic, and nature related research areas and applications. When fully polarimetric SAR data is not available, single- or dual-polarization SAR data can also be used whilst posing certain difficulties. For instance, traditional Machine Learning (ML) methods generally focus on finding more discriminative features to overcome the lack of information due to single- or dual-polarimetry. Beside conventional ML approaches, studies proposing deep convolutional neural networks (CNNs) come with limitations and drawbacks such as requirements of massive amounts of data for training and special hardware for implementing complex deep networks. In this study, we propose a systematic approach based on sliding-window classification with compact and adaptive CNNs that can overcome such drawbacks whilst achieving state-of-the-art performance levels for land use/land cover classification. The proposed approach voids the need for feature extraction and selection processes entirely, and perform classification directly over SAR intensity data. Furthermore, unlike deep CNNs, the proposed approach requires neither a dedicated hardware nor a large amount of data with ground-truth labels. The proposed systematic approach is designed to achieve maximum classification accuracy on single and dual-polarized intensity data with minimum human interaction. Moreover, due to its compact configuration, the proposed approach can process such small patches which is not possible with deep learning solutions. This ability significantly improves the details in segmentation masks. An extensive set of experiments over two benchmark SAR datasets confirms the superior classification performance and efficient computational complexity of the proposed approach compared to the competing methods.

An Unsupervised SAR Change Detection Method Based on Stochastic Subspace Ensemble Learning

As synthetic aperture radar (SAR) is playing an increasingly important role in Earth observations, many new methods and technologies have been proposed for change detection using multi-temporal SAR images. Especially with the development of deep learning, numerous methods have been proposed in recent years. However, the requirement to have a certain number of high-quality samples has become one of the main reasons for the limited development of these methods. Thus, in this paper, we propose an unsupervised SAR change detection method that is based on stochastic subspace ensemble learning. The proposed method consists of two stages: The first stage involves the automatic determination of high-confidence samples, which includes a fusion strategy and a refinement process; and the second stage entails using the stochastic subspace ensemble learning module, which contains three steps: obtaining the subsample sets, establishing and training a two-channel network, and applying the prediction results and an ensemble strategy. The subsample sets are used to solve the problem of imbalanced samples. The two-channel networks are used to extract high-dimensional features and learn the relationship between the neighborhood of the pixels in the original images and the labels. Finally, by using an ensemble strategy, the results predicted by all patches reclassified in each network are integrated as the detection result. The experimental results of different SAR datasets prove the effectiveness and the feasibility of the proposed method.

Seven years of surface deformation above the buried Nasr-Abad salt diapir using InSAR time-series analysis, Central Iran

This study employs Interferometric Synthetic Aperture Radar (InSAR) data sets to monitor the surface deformation of the Nasr-Abad buried salt diapir in the Central Basin of Iran. The Nasr-Abad salt diapir is one of the largest buried salt diapir in Iran and could be ideal site for oil/gas storage and industrial waste disposal. Forty ENVISAT Advanced Synthetic Aperture Radar (ASAR) images acquired during 2003–2010 were used to monitor surface activities of this diapir and its surrounding regions in that time window. One hundred and one interferograms were generated from both descending and ascending orbits by applying the Small-BAseline Subset (SBAS) method. Tropospheric artefacts in the displacement interferograms were mitigated using the power law correction method in TRAIN software. A time series of line-of-sight (LOS) displacements on the residual cap of Lower Red Formation above the buried diapir was also generated as well as the area. Finally, the data for temperature, precipitation, and tidal forces were correlated with the time-series displacement results of four points A, B, C, and D that located on the residual cap. Our analyses of our data indicate that surface above the diapir and an area of about 28 square kilometres subsided with maximum LOS velocity of about 7 mm/year for ascending images and 5 mm/year for descending images. Comparing the temperature and precipitation data with the time-series of displacement confirmed that they correlated with the same seasonal effects. Monitoring the activities of the Nasr-Abad region over 7 years shows that the region of surface subsidence is confined to the area that relates directly to the geological structures of the region.

Characterization of pre- and post-failure displacements of the Huangnibazi landslide in Li County with multi-source satellite observations

The Huangnibazi landslide in Li County of southwest China, began to slide rapidly since 8 August 2017. It received little attention before the failure event. The fast motion of the slope posed great threats to the safety of nearby villages and the national highway G317 at the foot of the mountain. In this study, Synthetic Aperture Radar (SAR) datasets collected by C-band Sentinel-1 and X-band TerraSAR-X (TSX), as well as optical images acquired by Sentinel-2 Multi-Spectral Instrument (MSI) over the landslide site were analyzed to characterize the evolution of the slope before and after the failure event. Firstly, the surface change of the sliding area was qualitatively evaluated using the Sentinel-2 true-color images and pseudo-colored TerraSAR-X amplitude images. Next, the slow movement of the slope before the failure event was detected by interferometry using both TerraSAR-X and Sentinel-1 data stacks. The maximum displacement rate estimated was larger than −45 mm/year along the radar line of sight (LOS) direction. Finally, the large post-failure displacements of the landslide were successfully retrieved with meter-level accuracy from TerraSAR-X observations using adaptive amplitude offset tracking method. The maximum displacement decreased from about 30 m in 11 days in August 2017 to 1–2 m in 22 days in January 2018 in both LOS and azimuth directions. According to the recent measurements of surface displacement, the landslide body is becoming stabilized gradually, which suggests that the risk of catastrophic collapse of the slope has been largely reduced.

On Extreme Winds at L-Band with the SMAP Synthetic Aperture Radar

In this letter, we discuss some observations of the Soil Moisture Active Passive (SMAP) mission’s high-resolution synthetic aperture radar (SAR) for extreme winds and tropical cyclones. We find that the L-band cross-polarized backscatter is far more sensitive to wind speed at extreme winds than the co-polarized backscatter and it is essential to observations of extreme winds with L-band SAR. We introduce a cyclone wind speed retrieval algorithm and apply it to the limited SMAP SAR dataset of cyclones. We show that the SMAP SAR instrument is capable of measuring extreme winds up to the category 5 (70 m/s) wind speed regime providing unique capabilities as compared to traditional scatterometers with C and Ku-band radars.

Surface deformation over the buried Nasr Abad salt diapir, Central Iran using interferometric synthetic aperture radar data

ABSTRACTThis study employs interferometric synthetic aperture radar (InSAR) data sets to monitor the surface deformation of the Nasr Abad buried salt diapir in the Central Basin of Iran. The Nasr Abad salt diapir is one of the largest buried salt diapirs in Iran and could be ideal site for oil/gas storage and industrial waste disposal. In this study, we use 40 advanced synthetic aperture radar (ASAR) images from Environmental Satellite (EnviSat) to analyse surface displacement rates of Nasr Abad diapir and its surrounding regions. A time series of line-of-sight (LOS) displacements on the residual cap above the buried diapir were obtained from both descending and ascending images between 2003 and 2010 by applying the Small-BAseline Subset (SBAS) technique. Tropospheric artefacts in the displacement interferograms were mitigated using the power-law correction method in Toolbox for Reducing Atmospheric InSAR Noise (TRAIN) software. Finally, the data for temperature, precipitation, and tidal forces were correlated with the time-series displacement results of four points that located on the residual cap. Our results indicate that surface above the diapir and an area of about 2 km2 subsided with maximum LOS velocity of about 7 mm year−1 for ascending images and 5 mm year−1 for descending images. The amount of vertical subsidence derived from LOS decomposition in reactive stage of Nasr Abad salt body is about 7 mm year−1. Comparing the temperature and precipitation data with the time series of displacement confirmed that the salt expands when temperatures increases and that salt flow accelerates when it is wet. The displacement pattern retrieved from InSAR analysis is in good agreement with intervals near maximum and minimum solid earth tides. Monitoring the activities of the Nasr Abad region over 7 years shows that the region of surface subsidence is confined to the area along the Khurabad and AbShirin-Shurab Fault zones in the southern Central Basin.

An Object-Based Assessment of Multi-Wavelength SAR, Optical Imagery and Topographical Datasets for Operational Wetland Mapping in Boreal Yukon, Canada

The authors evaluated multiple remotely sensed datasets for their contributions to operational wetland mapping in a subarctic, boreal cordillera study site in Yukon, Canada. They assessed Sentinel-2 optical imagery, Sentinel-1 C-band and ALOS PALSAR L-band synthetic aperture radar (SAR) imagery, and topographical data from the territorial digital elevation model (DEM) using an object-based image analysis (OBIA) approach. Three machine-learning algorithms were tested, namely random forest (RF), support vector machine (SVM) and k-nearest neighbor (KNN), using various data combinations (11 model scenarios). RF produced the most accurate results when incorporating all optical, SAR and DEM data (86.5%, kappa 0.84), with open water (100% producer accuracy, PA), marsh (75% PA) and swamps (85.7% PA) being detected most accurately. When assessed in isolation, Sentinel-2 optical data consistently generated more accurate classifications than either SAR platform or DEM data. RF variable importance metrics provided further explanation to these results, indicating the 8 most powerful variables to be optical. Variable reduction tests also produced comparable accuracies, indicating that an optimal RF model can be built based on predictive power rankings. The results can be used to inform resource managers on the efficacy of current datasets and their applications to wetland mapping in northern, subarctic environments.

Systematic preparation and processing of interferometric synthetic aperture radar data for monitoring linear transportation infrastructure

We propose a methodology for the systematic preparation and processing of interferometric synthetic aperture radar (InSAR) data for monitoring linear transportation infrastructure subject to geohazards. The methodology is applied to two RADARSAT-2 Spotlight synthetic aperture radar datasets, and three case studies in Cornwall, Eastern Ontario, Canada, are examined. An InSAR processing sequence was established and 19 SLA24 and 15 SLA74 images were used to create time-series deformation maps spanning from March 2015 to September 2016. The noise floors were ±1.5 and ±1.0 cm, for the SLA24 and SLA74 datasets, respectively. Phase unwrapping errors, atmospheric path delay, and the limited number of images were identified as the largest contributors to measurement uncertainty, which was of the same order as the ground deformation field. To improve coherence and utility of the radar images for monitoring the effects of geohazards on infrastructure, it is recommended that imagery acquisitions consider the use of small incidence angles with moderate image resolution and 6- to 12-day revisit periods.

A SAR Dataset of Ship Detection for Deep Learning under Complex Backgrounds

With the launch of space-borne satellites, more synthetic aperture radar (SAR) images are available than ever before, thus making dynamic ship monitoring possible. Object detectors in deep learning achieve top performance, benefitting from a free public dataset. Unfortunately, due to the lack of a large volume of labeled datasets, object detectors for SAR ship detection have developed slowly. To boost the development of object detectors in SAR images, a SAR dataset is constructed. This dataset labeled by SAR experts was created using 102 Chinese Gaofen-3 images and 108 Sentinel-1 images. It consists of 43,819 ship chips of 256 pixels in both range and azimuth. These ships mainly have distinct scales and backgrounds. Moreover, modified state-of-the-art object detectors from natural images are trained and can be used as baselines. Experimental results reveal that object detectors achieve higher mean average precision (mAP) on the test dataset and have high generalization performance on new SAR imagery without land-ocean segmentation, demonstrating the benefits of the dataset we constructed.

Multi-Feature Based Ocean Oil Spill Detection for Polarimetric SAR Data Using Random Forest and the Self-Similarity Parameter

Synthetic aperture radar (SAR) is an important means to detect ocean oil spills which cause serious damage to the marine ecosystem. However, the look-alikes, which have a similar behavior to oil slicks in SAR images, will reduce the oil spill detection accuracy. Therefore, a novel oil spill detection method based on multiple features of polarimetric SAR data is proposed to improve the detection accuracy in this paper. In this method, the self-similarity parameter, which is sensitive to the randomness of the scattering target, is introduced to enhance the discrimination ability between oil slicks and look-alikes. The proposed method uses the Random Forest classification combing self-similarity parameter with seven well-known features to improve oil spill detection accuracy. Evaluations and comparisons were conducted with Radarsat-2 and UAVSAR polarimetric SAR datasets, which shows that: (1) the oil spill detection accuracy of the proposed method reaches 92.99% and 82.25% in two datasets, respectively, which is higher than three well-known methods. (2) Compared with other seven polarimetric features, self-similarity parameter has the better oil spill detection capability in the scene with lower wind speed close to 2–3 m/s, while, when the wind speed is close to 9–12 m/s, it is more suitable for oil spill detection in the downwind scene where the microwave incident direction is similar to the sea surface wind direction and performs well in the scene with incidence angle range from 29.7° to 43.5°.