In classification of multispectral remote sensing image, it is usually difficult to obtain higher classification accuracy if only consider image's spectral feature or texture feature alone. In this paper ,we prese...
详细信息
In order to improve the classifier performance in semantic image annotation, we propose a novel method which adopts learning vector quantization (LVQ) technique to optimize low level feature data extracted from given ...
详细信息
Weakly-supervised semantic segmentation (WSSS) receives increasing attentions from the community in recent years as it leverages the weakly annotated data to solve the problem of lacking of fully annotated data. Among...
详细信息
In this paper, a novel image stitching method is proposed, which utilizes scale-invariant feature transform (SIFT) feature and single-hidden layer feedforward neural network (SLFN) to get higher precision of parameter...
详细信息
ISBN:
(纸本)9781509006212
In this paper, a novel image stitching method is proposed, which utilizes scale-invariant feature transform (SIFT) feature and single-hidden layer feedforward neural network (SLFN) to get higher precision of parameter estimation. In this method, features are extracted from the image sets by the SIFT descriptor and form into the input vector of the SLFN. The output of the SLFN is those translation, rotation and scaling parameters with respect to reference and registered image sets. We also apply a fast learning scheme, called pseudoinverse learning, to train SLFN to get higher training efficiency. Comparative experiments are performed between our proposed method and the traditional random sample consensus (RANSAC) based method. The results show that our method has the advantage not only at accuracy but also remarkably at fast speed.
Recently sparse coding with spatial pyramid matching method has shown its excellent performance in image classification. Inspired by this technique, we present an image classification approach by learning the optimal ...
详细信息
ISBN:
(纸本)9781467314886
Recently sparse coding with spatial pyramid matching method has shown its excellent performance in image classification. Inspired by this technique, we present an image classification approach by learning the optimal Multiple Pooling Combination strategy based on Non-Negative Sparse Coding (MPC-NNSC) in this paper. First, non-negative sparse coding with three different pooling methods as well as spatial pyramid matching method are utilized to encode local descriptors for image representation, respectively. Then a promising weight learning approach is employed to find a set of optimal weights for best fusing all these pooling methods in different scales. Lastly, support vector machine classifier with linear and histogram intersection kernel is employed for the final classification task. Experiments on two popular benchmark datasets are presented and they demonstrate the better performance of the proposed scheme compared to the state-of-the-art methods.
Recently, Discriminative Correlation Filter based trackers have increasingly become popular in the domain of visual object tracking, which is benefited by their effective and robustness in terms of tracking performanc...
详细信息
Kernel independent component analysis (KICA) has an important application in blind source separation, in which how to select the optimal kernel, including the kernel functional form and its parameters, is the key issu...
详细信息
ISBN:
(纸本)9781509006212
Kernel independent component analysis (KICA) has an important application in blind source separation, in which how to select the optimal kernel, including the kernel functional form and its parameters, is the key issue for obtaining the optimal performance. In practices, a single kernel is usually chosen as the kernel model of KICA in light of experience. However, selecting a suitable kernel model is a more difficult problem if one has not sufficient experience. To deal with this problem, an evolution based method to select the kernel model of KICA is proposed in this paper. There are two main features of the proposed method: one is that using a multiple kernel model, a convex combination of several single kernels, replaces the single kernel model;another is that particle swarm optimization (PSO) algorithm is utilized to find the combination weights of the composite kernel. Experiments conducted on separating one-dimensional mixed signals, nature images, and spectroscopic CCD images showed that using multiple kernels model with PSO kernel selection algorithm can enhance the performance of KICA.
image segmentation is the basis of imageprocessing and image analysis. However, there are no common method that can be used in natural images, and present methods fail to explain understandings of human's visual ...
详细信息
image segmentation is the basis of imageprocessing and image analysis. However, there are no common method that can be used in natural images, and present methods fail to explain understandings of human's visual system. In this paper, we propose to apply Karklin's visual perception model to extract feature vectors of images, and the features are clustered with K-means method. The results obtained in feature space are projected back to the image space to finish segmentation. A comparison with the Normalized Cuts (Ncut) method is done, and it turns out that proposed method outperform Ncut in texture rich images.
An adaptive SAR image enhancement method is presented for reducing the speckle noise and increasing the contrast of synthetic aperture radar (SAR) images. First, a fuzzy logic based filter, employing fuzzy edge to wei...
详细信息
A novel 3D terrain matching algorithm is presented in this paper. A terrain feature vector map (FVM), composed of local mean and local gradient, is employed to represent the terrain elevation map (TEM). Compared with ...
详细信息
ISBN:
(纸本)9780819469519
A novel 3D terrain matching algorithm is presented in this paper. A terrain feature vector map (FVM), composed of local mean and local gradient, is employed to represent the terrain elevation map (TEM). Compared with traditional matching algorithm using the magnitude of gradient to match, the new algorithm uses each component of the gradient vector to match individually, and it is able to generate two interim matching positions. Different from traditional matching algorithms which usually estimate an optimum matching position under some criterions at the end, the new algorithm fused the two interim matching positions to generate a final matching position or refuse to position in order to increase the matching confidence, which is very important because it is hardly acceptable to employ a mismatched position to correct the error of Inertial Navigation System (INS). Due to the stability of terrain and the high-precision of lidar ranging, the mean of a sensed terrain elevation map (STEM) sized terrain is quite stable. So it is bestowed to accelerate the matching process and to reduce mismatches at different terrain heights. Compared with other mismatch-eliminated methods based on neural network (NN) or support vector machine (SVM), the new method do not need training samples and is more stable and robust. Experimental results show that the proposed algorithm is effective and robust.
暂无评论