A new fingerprint matching method based on a dual-image template and an improved matching algorithm is presented in this paper. The proposed matching method is tested using images from FVC2002 fingerprint databases an...
详细信息
ISBN:
(纸本)9781538619377
A new fingerprint matching method based on a dual-image template and an improved matching algorithm is presented in this paper. The proposed matching method is tested using images from FVC2002 fingerprint databases and demonstrates a superior performance compared with fingerprint matching algorithm using single-image template. Compared with other dual-image matching algorithms, the proposed algorithm has advantages of easy implementation. By taking advantages of dual-image template and improved matching algorithm, the proposed matching method improves matching accuracy and relaxes requirement on fingerprint image quality.
The Hough transform has been widely used to determine the location and orientation of straight lines in images in the fields of computer vision and imageprocessing. The problem of the traditional Hough transform is t...
详细信息
ISBN:
(纸本)9781538619377
The Hough transform has been widely used to determine the location and orientation of straight lines in images in the fields of computer vision and imageprocessing. The problem of the traditional Hough transform is the large amount of calculation and inability to meet the requirements of real-time computing. In this paper, a new parallel Hough transform method which is based on GPU is presented to solve the problem. The method which uses the parallel computing accelerates the computing process and greatly improves the computational efficiency. Therefore, the imageprocessing calculations for linear features extraction process can achieve real-time requirements. The comparative experiment shows that the computing time can reach up to 3ms, and the speed-up ratio can be around 20 times.
Facial malformations such as overbite or underbite can be treated by maxillofacial surgery where the mandibula (i.e. the lower jaw) and/or maxilla (i.e. the upper jaw) is repositioned to obtain a preferential occlusio...
详细信息
Facial malformations such as overbite or underbite can be treated by maxillofacial surgery where the mandibula (i.e. the lower jaw) and/or maxilla (i.e. the upper jaw) is repositioned to obtain a preferential occlusion. These treatments need to be carefully planned prior to the actual operation. The planning of such surgery is an intensive and time consuming procedure in which the experience and knowledge of the surgeon plays a major role. As part of a CAD-based surgery planning, haptic interaction using 3D virtual models of the mandibula and maxilla, haptic feedback can make planning and surgery much more efficient. Haptic devices bridge the gap between the virtual environment and the physical world. An important step in such haptics based workflow is the collision detection between the two models at an update frequency of at least 1000 Hz. In this paper the possibility of the Implicit Sphere Tree algorithm as a collision detection algorithm tailored for this specific application is presented.
This paper presents a new magnetic resonance (MR) image reconstruction method which focuses on estimating the largest K wavelet coefficients. We model the MR image within union of subspaces framework and propose an al...
详细信息
ISBN:
(纸本)9781538619377
This paper presents a new magnetic resonance (MR) image reconstruction method which focuses on estimating the largest K wavelet coefficients. We model the MR image within union of subspaces framework and propose an algorithm named as Subspace Update Algorithm (SUA) to identify subspace. Then we estimate the value of the largest K coefficients by solving the optimization problem consisting of a data fidelity term and a total variation (TV) regularization term. Experimental results on practical MR brain scans show that our proposed method provides high quality reconstruction results. The approach can be applied in many different imaging scenarios.
Generative Adversarial Net is a frontier method of generative models for images, audios and videos. In this paper, we focus on conditional image generation and introduce conditional Feature-Matching Generative Adversa...
详细信息
ISBN:
(纸本)9781538619377
Generative Adversarial Net is a frontier method of generative models for images, audios and videos. In this paper, we focus on conditional image generation and introduce conditional Feature-Matching Generative Adversarial Net to generate images from category labels. By visualizing state-of-art discriminative conditional generative models, we find these networks do not gain clear semantic concepts. Thus we design the loss function in the light of metric learning to measure semantic distance. The proposed model is evaluated on several well-known datasets. It is shown to be of higher perceptual quality and better diversity then existing generative models.
This paper introduces a new set of orthogonal moment functions - U moments (UMs). The basis function of UMs is U-system, which is a new complete orthogonal piecewise polynomial function system. Compared with the tradi...
详细信息
ISBN:
(纸本)9781538619377
This paper introduces a new set of orthogonal moment functions - U moments (UMs). The basis function of UMs is U-system, which is a new complete orthogonal piecewise polynomial function system. Compared with the traditional orthogonal moments such as Legendre moments(LMs) and Zernike moments(ZMs), UMs has rather low computational complexity and have not numerical approximation errors, which benefit from its low-order and simple mathematical expressions. These properties ensure that UMs is a promising method in image reconstruction and real-time applications. Our experimental results indicate that UMs have a better image representation.
The probability of generating chaos can be substantially increased by cross-iteration between a class of special quadratic functions and random ones, and the result shows that it is a better method to construct chaoti...
详细信息
ISBN:
(纸本)9781538619377
The probability of generating chaos can be substantially increased by cross-iteration between a class of special quadratic functions and random ones, and the result shows that it is a better method to construct chaotic system. The paper works on the chaotic characteristic of the dynamic system that constitutes of a quadratic surface mapping and an image function in a spatial unit area, the chaotic attractor is discovered as a new characteristic for image recognition. The chaotic attractor changes with the parameters and surface control points. In particular, the number of the attractor points but the distributive distinction can be changed by slightly adjusting the surface or image, and it has important significance for research on pattern recognition or vision mechanism. In addition, the experimental result shows that using the image chaotic attractor to recognize different human faces is measurable.
The crucial issue in image quality enhancement is the fidelity issue. Three fidelity criteria (3FC) are proposed in image quality enhancement. FC 1 is that the information entropy of the enhanced image should not be b...
详细信息
ISBN:
(纸本)9781538619377
The crucial issue in image quality enhancement is the fidelity issue. Three fidelity criteria (3FC) are proposed in image quality enhancement. FC 1 is that the information entropy of the enhanced image should not be bigger than the original image. FC 2 is that the constituents of the enhanced image should not be bigger than the original. FC 3 that the color relationship in the enhanced image is not changed when comparing with the original. Our studies point out that the image enhancement methods based on histogram equalization and Retinex do not meet three fidelity criteria in image quality enhancement.
In this paper we present a study on historical documents denoising methods and make visual quality performance comparison through Deep Residual Learning, Alternating Direction Method of Multiplier and Anisotropic diff...
详细信息
ISBN:
(纸本)9781538619377
In this paper we present a study on historical documents denoising methods and make visual quality performance comparison through Deep Residual Learning, Alternating Direction Method of Multiplier and Anisotropic diffusion PDE. Experimental results demonstrate their denoising visual quality performance and we make a comparison to in different condition respectively.
The traditional method of mine belt conveyor fire accident detection is limited. A novel imageprocessing method is proposed in this paper, which combines 2D-Dimensional Otsu, Canny dge detection and Artificial Bee Co...
详细信息
ISBN:
(纸本)9781538619377
The traditional method of mine belt conveyor fire accident detection is limited. A novel imageprocessing method is proposed in this paper, which combines 2D-Dimensional Otsu, Canny dge detection and Artificial Bee Colony algorithm. The noise interference of image is reduced by using median filtering techniques. In order to obtain gray value and neighborhood gray scale average, 2-Dimensional histogram of image is constructed. The fitness function of Artificial Bee Colony algorithm is designed by 2-Dimensional Otsu method and the maximum value of the fitness function is the optimal threshold for image segmentation. Finally, the canny edge detection and grayscale morphology were used to extract the target. Theoretical analysis and simulation results show that the proposed method is effective to detect fire accident of mine belt conveyor in complex undeground environment.
暂无评论