2005 Vol. 30, No. 8
A method of the block adjustment based on a new strict geometric model for remote sensing images with high resolution (RSIHR), using few control points, is presented in this paper. The model has been tested for IKONOS and other RSIHR. The block adjustment based on RPC parameters with IKONOS images has been tested and compared with the new strict geometric model. All the tested result shows that the accuracy of the new model can reach the level corresponding to the ground resolution of the images with few control points.
A new method using multidimensional change template analysis guided by GIS knowledge is proposed in this paper. The new method combine the use of quad-tree change detection template and GIS knowledge. It has been used to detect change among different temporal digital orthophoto maps (DOM) and then update image database. Test results show that this method is effective reliable and robust in many cases.
In this paper, the shape of contour line is studied with geometry theory. It is concluded that the shape of neighbor contours change similarly. This similarity reflects the terrain surface changing within the neighbor contours. A new method for reconstruction of terrain surface is put forward with similarity rule. It bases on the theory of potential field in physics. The new method can make full use of the information in original data. It can overcome some shortcomings of traditional method. The experimental results show that the terrain surface reconstructed by the new method contains more detail information than that by traditional method, and it can reflect the real terrain variations within the neighbor contours.
The paper mainly introduces how to get the parallel image from the lean image using the conditions of geometry and restriction in the surface of building according to the theory of perspective transformation. After transformation, the principal distance of rectified image has the same value as one of original image. Then, using the least control condition, the parallel images can be magnified to the scaled image. The technology proposed in paper made getting facade image with large scale rather rapidly, not influenced by the height of building, and can be used in many kinds of mobile mapping situation.
In this paper, 152 suspended sediment concentration resulting images retrieved from reflectance of NOAA/AVHRR channel 1 (580 nm~690 nm) and channel 2 (720 nm~1 100 nm) in the Pearl River estuary and its adjacent coastal waters from 1995 to 2000 were selected to analyze the spatial distribution and variation of suspended sediment concentration in three temporal scales. The modeler of ERDAS IMAGINE ver8.5 was employed to calculate pixel based average concentration of suspended sediment according to the special time period and the regulation of spatial and temporal distribution and variation of suspended sediment concentration was studied. Results showed that obvious spatial distribution and temporal variation existed in all three temporal scales, concentration contours almost paralleled coastline and the values decreased from river mouth to open sea. The curvature and gradient implied the impact of run-off on suspended sediment along the coastal waters. The shift and convex of concentration contour basically directed the impact scope of run-off and the movement of suspended sediment along the coastal waters.
This paper deals with the airport recognition from optical remote sensing images under the framework of feature level information fusion by using BP neural network. Automatically extracting airport from remote sensing imagery is one of research hots which has being attracted much research interests both from the computer vision community and remote sensing application field. Due to its uniqueness in geometric shape, shape property analysis on airport and recognition based on shape parameters or directly on shape recognition is the intutitive approch in order to recognize airport. However, though straitfordward in method, recognition based on shape analysis only can cause unavoidalbe mistakes when there are objects similar in shape around the airport such as express ways.Thanks to the multispectral property of most commecial remote sensing images, this paper expound an approach for airport recognition based on multi feature fusion, namely, shape parameter, spectral signature and texture feature of an image. Since all these features are different in unit and order of magnitude, it is difficult to combine these features in commonly used algorithms, such as Bayesian inference. Error Back Propagation Neural Network is used to fuse all these features. Two revised version of BP network is used in order of improving the convergence speed during training and recognizing. Real TM images are used for case study, showing the effectiveness of our approach.
Different routes are designed and applied to extract the two kinds of ships in this paper. The method can be divided into three main steps: extract water polygons with histogram segmentation, extract holes in the water polygons with morphological operations as the possible sailing ships, and extract possible moored ship with the identification of salients along the water boundary, and then screen real ships out of these possible ships with more shape constraints. Experiments give good results.
This paper proposes an effective approach to extract linear object in high-resolution remote sensing image. The approach integrates the knowledge about the linear object to implement the watershed algorithm and guide the region merging and finally extract the linear object. First, the Kalman filter algorithm is used to detect the straight line in the image, the center point of parallel line pairs and the minimum with dynamics larger than predefined threshold are utilized as marker point to modify the morphological gradient of the input image by geodesic reconstruction, the modified gradient image is then segmented by the watershed transform. The initial segmentation result is input to region merging process. This process applies the region adjacency graph (RAG) representation of the segmented regions and knowledge about the road to execute the region merging, which significantly reduce the merging time. The proposed scheme was tested on remote sensing images of 2 m resolution, and the results show that the extraction of road is quite promising.
SVM has desirable classification ability even if with fewer samples. In addition, LS-SVM reduces the complexity further through replacing the inequality of SVM by equality. This paper applies LS-SVM to aerial images segmentation. This paper researches on the different kernel and sparse of LS-SVM. The kernel influences aerial image segmentation. The briefness of decision function is reached by the LS-SVM sparseness. The experiments show the segmentation results of LS-SVM are better.
VIUPD is a new vegetation index based on the UPDM, which uses almost the whole vegetation information obtained by each sensor. It has been verified that the VIUPD is sensitively to reflect the amount of vegetation and the degree of vegetation vigor by using ground-measured data. This paper give an introduction of calculating the VIUPD, the results demonstrate that our conclusions are correct by comparing it with NDVI and EVI.
In this paper, the authors introduce the theory base of split-window algorithm and make a derivation for the split-window algorithm. As to the characters of the MODIS bands, the authors make a simplification for the Planck function. For the two main parameters of the split algorithm, the authors obtain them from the same image. Finally the authors retrieve the LST of Huanbohai region by using MODIS data and get the spatial distribution of LST in this region and make some analysis for the retrieval results.
In this paper, the authors introduce a new remote sensing monitoring methed. The monitored data are analyzed synthetically. These results can provide abundant parameters and validated data to GCM (The Globe Climate System Model). Simultaneously, it provide a kind of new remote sensing technology to long-term forecasting on Asia dust storms.
A practical localization formula is given in the paper on the basis of the principle of airborne SAR imaging. This method is used to rectify the 0.5m airborne image and the result is satisfactory. The error of SAR image localization is also proposed according to the error propagation law and some conclusions are drawn based on some error simulations.
On the basis of the research of periodic median algorithm, a new adaptive algorithm is present. The moving mask average with adaptive weight is conducted, which is based upon the gradient of phase. The less of the weight is corresponding to the larger gradient value so that the boundary of the fringe is not blurred. Additionally , the moving mask is small enough to maintain the continuities of the fringes. Comparing with existing approach for phase noise filtering , the effectiveness of the proposed approach is verified with the interferogram generated from ERS-1/2 InSAR data.
The image of atmospheric degradation has the most important characteristic: the degree of influence is correlative with the distance between scene and camera. Pointing at the correlation, a new effective approach based on character of images of atmospheric degradation is brought forward. The experimental result show clearly that the approach deal well with images of atmospheric degradation and does not need any secondary information.
In the first section, according to the assumption that only the silhouette could be extracted to describe the motion of a moving object, a three-frame differencing model is constructed based on the variance ratio of successive blocks of different frames, simultaneously the Gaussian kernel based density estimation model in \ is used as a filter to suppress false positives, and the SVC algorithm in \ is also employed to realize real-time multi-object motion detection. A novel contour tracing algorithm, namely the RW (roller wheel) method, is proposed in the next section. Edge information detected by the forgoing model can be fully used by this algorithm, while the interior details will be skipped. This algorithm works in a similar way that human vision system works. Experimental results are given in the end.
Based on the combinational representation of spatial topological relations, how to combine basic spatial topological relations is analyzed in this paper. According to the combinational method, two kinds of systems of spatial topological relations between two areas are given, in addition, a reasoning table about spatial topological relations is given, and a complete diagram about spatial topological relations between two areas is listed.
The traditional principal component analysis was improved from the three aspects such as lineal transform of negative index, logarithmic treatment of original index and choosing the evaluation index. Then it was applied to the water pollution evaluation in East Lake. Comparing with the traditional methods, it could hold more information of the original index matrix and had much higher contributing rate of the first principal component. The whole process was accomplished by the computer and created a water pollution map.
In this paper, the formulae of local parameters are firstly derived under the network-solution and the PPP-solution conditions, respectively. If the weight matrix of global parameter in PPP-solution is small enough, the expression of local parameters are the same as that in network-solution. Then, 16 daily solutions are obtained in both PPP mode and network mode. Three sites are selected to compare their solution differences. The experimental results demonstrated that their difference between two solutions in coordinates and tropospheric delays are only few millimeters.
In this paper, the methods of ERP parameters estimation based on IGS SINEX file of GPS solution are discussed in details. To estimate ERP parameters, two different ways are involved: one is the parameter transformation, the other is direct adjustment method with restrictive conditions. And the estimated results with independent copyright program are compared to IERS results.
Basing on the Gauss-Markov model and according to the basic functional model this paper distinguishes the different phase ambiguity in DEM generation and deformation extraction by establishing the phase unwrapping functions. Thereafter a refined D-InSAR functional model is proposed . According to the refined D-InsAR functional model some basic D-InSAR models have been derived, such as two-pass, three-pass and four-pass method. Some concise error analyses have also been give during the derivation.
Based on satellite gravity and geoid observation results, we calculated and analyzed different order and degree of harmonic function of geoid, which can express earth shape. It is found that 26 degree geoid anomaly shows the fundamental shape of earth shape asymmetry, and geoid anomaly of more than 6 degree only shows the local undulation character of the earth shape. Moreover, The authors inverted three dimensional global mantle density (anomaly) combining geoid data and seismic tomography data by means of damping least square method. The density anomaly result shows that mantle density is obviously inhomogeneity not only in horizontal direction, but also in radial direction. Comparing low degree geoid anomaly with mantle density anomaly, The authors think the earth shape asymmetry is mainly caused by lower mantle density inhomogeneity.