An efficient method for facial feature location was proposed based on skin color segmentation and gray image symmetry transform. To reduce the influence of background and variable facial poses, the Blob analysis and e...
详细信息
An efficient method for facial feature location was proposed based on skin color segmentation and gray image symmetry transform. To reduce the influence of background and variable facial poses, the Blob analysis and ellipse fitting methods were successively applied to the image blobs detected by conventional skin color segmenting method. Invalid blobs were removed and candidate face regions were adjusted vertically. Then, mouth region was extracted from each candidate face region by segmenting hue image. Afterwards, the generalized symmetry transform (GST) of gray image was used to obtain candidate eyes and the real eyes were detected by combination optimization of the proposed cost function which was related to model of facial geometry and positions of eyes and mouth. Finally, other facial features such as chin and cheek could be precisely located. The test results show that the proposed approach is more effective and has better performance.
Instead of traditionally using a 3D physical model with many control points on it, a calibration plate with printed chess grid and movable along its normal direction is implemented to provide large area 3D control poi...
详细信息
Instead of traditionally using a 3D physical model with many control points on it, a calibration plate with printed chess grid and movable along its normal direction is implemented to provide large area 3D control points with variable Z values. Experiments show that the approach presented is effective for reconstructing 3D color objects in computer vision system.
This paper introduces a new class of switching vector median filter. The proposed algorithm first uses four directional masks to analyze the color difference between the central pixel and its neighborhood pixels in th...
详细信息
ISBN:
(纸本)9780819469502
This paper introduces a new class of switching vector median filter. The proposed algorithm first uses four directional masks to analyze the color difference between the central pixel and its neighborhood pixels in the RGB color space and classify each color pixel into noisy pixel or noise-free one, and then employs the standard vector median filtering operations in the detected noisy locations to restore the corrupted pixels and leave the noise-free ones unchanged. The simulation results show that the proposed method excellently suppresses impulsive noise as well as preserving the image details well, and significantly outperforms the existing vector filtering solutions in terms of both the objective measures and the perceptual visual quality.
In this paper, a new genetic algorithm for solving multi-constrained optimization problems based on KS function is proposed. Firstly, utilizing the agglomeration features of KS function, all constraints of optimizatio...
详细信息
In this paper, a new genetic algorithm for solving multi-constrained optimization problems based on KS function is proposed. Firstly, utilizing the agglomeration features of KS function, all constraints of optimization problems are agglomerated to only one constraint. Then, we use genetic algorithm to solve the optimization problem after the compression of constraints. Finally, the simulation results on benchmark functions show the efficiency of our algorithm.
While null space based linear discriminant analysis (NLDA) obtains a good discriminant performance, the ability easily suffers from an implicit assumption of Gaussian model with same covariance each class. Meanwhile, ...
详细信息
The principle of target tracking and data fusion techniques are discussed. To resolve high uncertainty that exists in sensors of mobile robots, the cross-sensor and cross-modality (CSCM) data fusion algorithm is prese...
详细信息
ISBN:
(纸本)9781424417612;1424417619
The principle of target tracking and data fusion techniques are discussed. To resolve high uncertainty that exists in sensors of mobile robots, the cross-sensor and cross-modality (CSCM) data fusion algorithm is presented. The algorithm is based on particle filter techniques, fuses the information coming from multiple sensors and merges different state space models. So it can be used to eliminate system and measurement noise and estimate value of position and heading of mobile robot. On simulation experiments, we compare different cases such as single sensor and multi-sensor data fusion, the results demonstrate the feasibility and effectiveness of this algorithm and exhibits good tracking performance.
Panchromatic (Pan)-sharpening of multispectral (MS) bands is an important technique in various applications of satellite remote *** this paper, we apply the support value transform (SVT) to Ikonos image *** fused sali...
详细信息
Panchromatic (Pan)-sharpening of multispectral (MS) bands is an important technique in various applications of satellite remote *** this paper, we apply the support value transform (SVT) to Ikonos image *** fused saliency features are represented by support values and extracted by *** low-resolution MS bands are resampled to the fine scale of the Pan image and sharpened by injecting the detailed features extracted from the high-resolution Pan *** fusing results on Ikonos MS + Pan data demonstrate that the proposed image fusion method is effective and efficient.
Panchromatic (Pan)-sharpening of multispectral (MS) bands is an important technique in various applications of satellite remote sensing. In this paper, We apply the support value transform (SVT) to Ikonos image fusion...
详细信息
ISBN:
(纸本)9781424409723
Panchromatic (Pan)-sharpening of multispectral (MS) bands is an important technique in various applications of satellite remote sensing. In this paper, We apply the support value transform (SVT) to Ikonos image fusion. The fused saliency features are represented by support values and extracted by SVT. The low-resolution MS bands are resampled to the fine scale of the Pan image and sharpened by injecting the detailed features extracted from the high-resolution Pan image. The fusing results on Ikonos MS + Pan data demonstrate that the proposed image fusion method is effective and efficient.
MRl simulation is a suitable method to evaluate and acquire understanding of the complex blood oxygenation level dependent (BOLD) effect. To study the observability of BOLD signal, we extended a MRI simulator. This si...
详细信息
MRl simulation is a suitable method to evaluate and acquire understanding of the complex blood oxygenation level dependent (BOLD) effect. To study the observability of BOLD signal, we extended a MRI simulator. This simulator is capable of complete modeling of object (tissue) and main field inhomogeneities. We made a digital phantom in which the magnitude of fractional oxygenation (Y), hematocrit (Hct) values in red blood cells, the diameter of vessels, and also the position of blood vessel within the voxel can be changed. Then we created echo planar imaging (EPI) images of the phantom under resting and activated states using the simulator. Resulting images were analyzed using cross-correlation and t-test analysis for activation detection. Then the effect of various parameters on the detection of BOLD is investigated. These parameters include fractional oxygenation, spatial extent of BOLD effect and the location of vessel inside voxel. This simulation study provides a basis for planning experiments aimed at BOLD contrast detection with EPI pulse sequence.
An improved adaptive wavelet thresholding method for image denoising was proposed to overcome the limitation of Donoho's VisuShrink and Lakhwinder Kaur's NormalShrink. According to the different sub-band chara...
详细信息
An improved adaptive wavelet thresholding method for image denoising was proposed to overcome the limitation of Donoho's VisuShrink and Lakhwinder Kaur's NormalShrink. According to the different sub-band characteristics, a new scale parameter equation was defined based on Lakhwinder Kaur's NormalShrink threshold, which was employed to determine the optimal thresholds for each step scale. Experimental results on several testing images show that the proposed method separates signals from noise completely in each step scale and eliminates white Gaussian noise effectively. In addition, the method also preserves the detailed information of the original image well, obtain superior quality image and improves Peak Signal to Noise Ratio (PSNR). Furthermore, since this method can improve the efficiency of image denoising and doesn't increase time complexity, it could be applied in the real-time processing.
暂无评论