Huang Shunji

Multi-target Imaging Processing Algorithms of ISAR Based on Time-frequency Analysis

This paper analyzed the features of the Doppler frequency shift of multi-target of ISAR at different moving states. Based on the model of rigidity multi-target, several modified time-frequency analysis approaches were proposed to simulate ISAR imaging of multi-target. The results showed the feasibility of the ISAR imaging method and the ability of recognition for multiple targets

Satellite-borne SAR quick-look processing algorithm

This paper puts forward a quick-look algorithm for reduced resolution SAR imaging processing, an analysis of the algorithm principle, the FIR filter choice, a correction method for the filter phase and altitude errors, addition modes on the azimuth direction, and the correction of the distance movement error and azimuth spectrum center. Experiments with satellite-borne SAR raw-data show that this algorithm needs a very small calculation time and a small memory volume to get good quality reduced-resolution SAR images.

Fusion of images with different spatial resolution

Nowadays in the field of information processing, it is not adequate for some tasks to use only one data set obtained with a single sensor. In this paper, the authors take advantage of pixel-level fusion which consists of merging information from different digital images, and concentrate on the fusion of two different spatial resolution images of the same scene obtained by SAR and optical sensors respectively. The fusion of the two image intensities is carried out by obtaining the image which has the least mean square error (LMSE) from the image with the finer resolution, subject to some constraints. The image achieved is of higher quality than the reconstruction obtained using only one image. The authors obtained more useful and detailed information which shows the effectiveness of their proposed method.

Development of an automatic target detection and characterisation system in polarimetric SAR images

An automatic algorithm that enables processing multi-look polarimetric SAR image data to automatically detect stationary ground man-made targets in diverse natural environments and then to characterise them by polarimetric characteristics is presented. The algorithm consists of three main building blocks: scaler CFAR, polarimetric discrimination, and unsupervised polarimetric classifier. The algorithm is applied to the polarimetric SAR images of San Francisco Bay for detection and classification of building targets.

The analysis of interferometric SAR imaging precision in the distributed micro-satellite system

The INSAR imaging is one of the principal working patterns of the distributed satellite SAR system. This paper research mainly the influence of the Earth-movement and the distributed satellite circling to the INSAR imaging precision, analyze and derive the corrected imaging formula and the error curvature, and analyze in detail the height error shift in the variant latitudes and rotating angles. The results of simulation verify that when the latitude is lower and the rotating value is larger, the height error is higher. And the height error is the highest near the equator.

An algorithm for integrating GPS/INS attitude determination system

The integration of inertial navigation system (INS) with Global Positioning System (GPS) receivers has been an area of intensive research in recent years. Using such an integrated system, precise attitude of moving platforms can be obtained. The discussion in this paper focuses on precise attitude determination by GPS/INS integration. A highly accurate attitude determination algorithm of a linear convex combination of a linear-two-point and a quadratic-five-point polynomial to filter and predict GPS/INS attitude data is presented in this paper. An integration experiment of real GPS data and simulated INS data shows the validity of this method.

A new method for edge detection of SAR image

A combined method for edge detection of SAR images is put forward in this paper. This method is composed of the wavelet transform, fuzzy median filter, adaptive thresholding and differential gradient operator. In using the fuzzy median filter, this paper gives a new method for adaptively selecting the factors of the fuzzy median filter. The experimental results show the effectiveness and stability of this method.

The Global Road Extraction Approach from Synthetic Aperture Radar Images

The extraction of road networks from synthetic aperture radar imagery is of fundamental importance for geospatial applications. The methods basically go through the steps such as road sharpening, road finding, road tracking and road linking. In this paper, a new global approach to road detection that is motivated by Kohonen neural network is proposed. It makes the joint of the roads more fluent and reduces detection errors. The computer simulation results show the highest extraction quality from the real SAR images.

Study of along track interferometric technology for the bistatic SAR

Bistatic synthetic aperture radar refers to a SAR system that the transmitter and receivers are placed in different platform. Because Bistatic SAR effectively avoids many shortcomings of conventional SAR, it becomes the research hot spot in the world of late years. Thereamong ATI is an important working model. In our work, we first analyze and research the imaging model of ATI in Bistatic SAR, then derive the velocity formula of the moving target. Next, we analyze the interferometric result in the case of arbitrary transmitter and receiver platforms. At last, we verify the accuracy of the formula by using the simulation and actual data, and declare the availability of the model. The result shows that the ATI technology can measure the radial velocity of the slowly moving target in Bistatic SAR system. If the direction of the baseline is arbitrary, by modifying the formula, we can get well result equally. Keywords-Bistatic SAR; ATI ;along-track interferometry; radial velocity; baseline; phase;

Automatic segmentation of video object planes in MPEG-4 based on spatio-temporal information

Segmentation of semantic Video Object Planes (VOP’s) from video sequence is a key to the standard MPEG-4 with content-based video coding. In this paper, the approach of automatic Segmentation of VOP’s Based on Spatio-Temporal Information (SBSTI) is proposed. The proceeding results demonstrate the good performance of the algorithm.Segmentation of semantic Video Object Planes (VOP’s) from video sequence is a key to the standard MPEG-4 with content-based video coding. In this paper, the approach of automatic Segmentation of VOP’s Based on Spatio-Temporal Information (SBSTI) is proposed. The proceeding results demonstrate the good performance of the algorithm.