Discretization of continuous attributes is one of the important steps in preprocessing of data analysis. In this paper, a new method of supervised discretization of continuous attributes based on entropy and hierarchi...
详细信息
Discretization of continuous attributes is one of the important steps in preprocessing of data analysis. In this paper, a new method of supervised discretization of continuous attributes based on entropy and hierarchical clustering guiding by level of consistency of decision table is introduced. This method makes use of the concept of the level of consistency of decision table in Rough Sets. According to the level of consistency of the produced decision table, the number of hierarchical cluster is adjusted dynamically in the first step. And then in the second step, we merge adjacent region based on entropy without damaging the level of consistency. Experiments show that this method is feasible.
Retinex method mainly consists of two steps: estimation and normalization of illumination. How to extract the background illumination accurately is a key problem. The backgrounds of picture sequence in video's adj...
详细信息
Retinex method mainly consists of two steps: estimation and normalization of illumination. How to extract the background illumination accurately is a key problem. The backgrounds of picture sequence in video's adjacent frames are usually similar and closely related. More accurate illumination information can be extracted when this characteristics of video's picture sequence is considered. In the paper, we propose an improved Retinex algorithm. Filter the images using the Gauss masks of different scale and parameter for each frame image, and all these filtering results are fused together by minimum method. In the paper, the scale of Gauss filters are set as 5, 9, 13, 25, and their variance set as 0.3, 0.5, 0.7 and 1.0 respectively. 6 adjacent frame images are selected, and the uniform and optical background image for these 6 images can be extracted by maximum method. This method makes use of the similarity and relationship among the adjacent frame images in videos. Enhance the images using Retinex method with this optical background image as their uniform illumination information. Experiment shows that more accurate back grounds are acquired and more excellent enhancement performance are achieved.
The present paper reports studies on thermodynamic quantities like temperature of the universe, heat capacity and squared speed of sound in f(R) and Horava-Lifshitz gravity theories. Considering the universe to be fil...
The present paper reports studies on thermodynamic quantities like temperature of the universe, heat capacity and squared speed of sound in f(R) and Horava-Lifshitz gravity theories. Considering the universe to be filled with dark matter and dark energy we have shown that in all cases the equation of state behaves like quintessence. The thermodynamic quantities have been studied graphically by plotting them against redshift z.
The nonlinear response and strong coupling of control channels of micro machined membrane deformable mirror (MMDM) devices make it difficult to obtain the desired optimal surface shape of the MMDM. In order to overcom...
详细信息
The nonlinear response and strong coupling of control channels of micro machined membrane deformable mirror (MMDM) devices make it difficult to obtain the desired optimal surface shape of the MMDM. In order to overcome these limitations, the wave front modal reconstruction algorithm based on particle swarm optimization (PSO) is proposed, which is used to find the optimal control voltages of the MMDM. It is initialized with a population of random solutions and searches for optima by updating generations. In this way the objective function can be optimum and influence of phase aberration can be reduced to minimum. The simulation results show that this algorithm could find the optimal MMDM shape which was applied to correct phase aberration.
Wavefront sensor-less adaptive optics is an important method for correcting wavefront aberration. An appropriate optimization algorithm is crucial to aberration correction. An improved particle swarm optimization (IPS...
详细信息
Wavefront sensor-less adaptive optics is an important method for correcting wavefront aberration. An appropriate optimization algorithm is crucial to aberration correction. An improved particle swarm optimization (IPSO) is presented which well addresses slow convergence rate and low calculation precision in the standard PSO. In IPSO, mutation strategies are adopted to keep the diversity of population and make particles explore the solution space efficiently. Based on these two algorithms, an adaptive optics system with a 37-element deformable mirror is established. The experimental results show that IPSO is better than SPSO in convergence rate, reconstruction performance and correction effect.
In order to eliminate noise in wavefront slope signals detected by a Hartmann wavefront sensor, a new denoising method was proposed by combining wavelet transform with modal reconstruction algorithm. The noise in the ...
详细信息
In order to eliminate noise in wavefront slope signals detected by a Hartmann wavefront sensor, a new denoising method was proposed by combining wavelet transform with modal reconstruction algorithm. The noise in the signals was removed by wavelet transform. The residual noise then was removed from the signals by using modal reconstruction algorithm. By practical application and simulation test, it was proved that this method could eliminate noise efficiently from wavefront slope signals during wavefront detection. The method that combines wavelet transform and modal reconstruction algorithm can be used for wavefront slope signals denoising.
The object-oriented simulation software of an adaptive optics system is used to evaluate the performance of the adaptive optics system. The software includes distorted wavefront correction and turbulence simulation. A...
详细信息
The object-oriented simulation software of an adaptive optics system is used to evaluate the performance of the adaptive optics system. The software includes distorted wavefront correction and turbulence simulation. An object-oriented adaptive optics system framework is proposed based on the features of adaptive optics systems simulation. Numerical simulation results show that it is a valid numerical simulation software.
This paper presents a novel scheme for face recognition by fusing local and global discriminant features. It has been observed that facial changes are occurred due to variations in facial expression, illumination cond...
详细信息
This paper presents a novel scheme for face recognition by fusing local and global discriminant features. It has been observed that facial changes are occurred due to variations in facial expression, illumination condition, pose, etc. and these changes are often appeared only some regions of the whole image. The global features extracted from the whole image are not able to cope with these facial changes. To cope with the above facial changes face images are divided into a number of non-overlapping smaller sub-images and discriminant features are extracted from these sub-images as well as from the whole image. All these extracted local and global features are fused to form a large feature vector. We have used generalized two-dimensional fisher's linear discriminate (G-2DFLD) method to extract these local and global discriminant features. We have used the fisher's linear discriminate (FLD) method to extract lower dimensional discriminant features from the fused large feature vector. A Multi-class Support Vector Machine (SVM) is applied on these reduced feature vector for classification. The proposed method was evaluated on AT&T Face Database and experimental results show that the performance of the proposed method is better than other global feature extraction methods like PCA, 2DPCA, PCA+FLD, 2DFLD and G-2DFLD methods.
Discretization of continuous attributes is one of the important steps in preprocessing of data analysis. In this paper, a new method of supervised discretization of continuous attributes based on entropy and hierarchi...
详细信息
Discretization of continuous attributes is one of the important steps in preprocessing of data analysis. In this paper, a new method of supervised discretization of continuous attributes based on entropy and hierarchical clustering guiding by level of consistency of decision table is introduced. This method makes use of the concept of the level of consistency of decision table in Rough Sets. According to the level of consistency of the produced decision table, the number of hierarchical cluster is adjusted dynamically in the first step. And then in the second step, we merge adjacent region based on entropy without damaging the level of consistency. Experiments show that this method is feasible.
暂无评论