A new SAR image de-noising approach that uses the local properties analysis of SAR image and the discrete non-separable shearlet transform (DNST) is proposed in this paper. According to the local properties analysis m...
详细信息
ISBN:
(纸本)9781538619377
A new SAR image de-noising approach that uses the local properties analysis of SAR image and the discrete non-separable shearlet transform (DNST) is proposed in this paper. According to the local properties analysis method, the SAR image is divided into homogeneous region, non-homogeneous region and target region. The homogeneous region uses the average filter to de-noising. The non-homogeneous region uses the DNST transform to de-noising and target region is reserved directly. The experimental results show that the proposed approach can efficiently reduce the speckle noises and improve the edge-preserving ability.
Recent years, the efficiency and accuracy of content based on image retrieval (CBIR) are two research directions. In order to improve retrieval precision, firstly, we use a fusion framework, combined with Color Differ...
详细信息
ISBN:
(纸本)9781538619377
Recent years, the efficiency and accuracy of content based on image retrieval (CBIR) are two research directions. In order to improve retrieval precision, firstly, we use a fusion framework, combined with Color Difference Histogram (CDH) and Micro-structure Descriptor (MSD) which are two kinds of image descriptors to calculate distances between all images and transform the distances between images into similarities. Then we utilize the similarities to construct a similarity matrix. Secondly, by using the similarity matrix, we construct a hypergraph, the hypergraph combined with relevance feedback technique is exploited to rank images and get similar images. Extensive experiments are carried out on the Corel-5K dataset and get a very good performance.
Multilayer image segmentation is commonly used in computer aided diagnosis and therapy planning. However, manual multilayer image segmentation is time-consuming and tedious. In this paper, an efficient and accurate se...
详细信息
ISBN:
(纸本)9781538619377
Multilayer image segmentation is commonly used in computer aided diagnosis and therapy planning. However, manual multilayer image segmentation is time-consuming and tedious. In this paper, an efficient and accurate semi-automatic method, which is based on Gaussian weighted Euclidean distance and nonlinear interpolation, is presented. The first and the last layers are segmented using improved live wire method. Then the prior knowledge, which is from the Gaussian weighted Euclidean distance transformation of the first and last layers, is combined with nonlinear interpolation to realize automatic segmentation of the intermediate layers. Experiments were conducted over magnetic resonance (MR) images of human leg. The results show that proposed method can not only reduce the time consumption, but also improve the accuracy of the segmentation.
In this paper, an improvement of the Canny edge-based image expansion algorithm is proposed. Our new expansion algorithm preserves the edges of an object. It generates the higher contrast and sharper images through mo...
详细信息
ISBN:
(纸本)9781538619377
In this paper, an improvement of the Canny edge-based image expansion algorithm is proposed. Our new expansion algorithm preserves the edges of an object. It generates the higher contrast and sharper images through modification of the neighborhood pixel values of the edges. In this method, we define two cases according to the orientations of the edges. In any diagonal orientation, we define two new operators to determine whether the diagonal direction is the left or right diagonal. For different cases, we propose different functions to process the neighborhood pixel values of the edges. Finally, we compare the expansion results and analyze the resulted image contrasts from different expansion algorithms. Our proposed expansion method generates higher contrast and less blurring and zigzag images with crisper appearances.
FCM algorithm is a popular algorithm for medical image segmentation. The precise process of segmenting brain tissue images becomes more challenging in the presence of noise and other image artifacts. An improved adapt...
详细信息
ISBN:
(纸本)9781538619377
FCM algorithm is a popular algorithm for medical image segmentation. The precise process of segmenting brain tissue images becomes more challenging in the presence of noise and other image artifacts. An improved adaptively regularized kernel FCM method is proposed in this paper. The spatial constraint function of membership is introduced to enhance clustering by adjusting the degree of influence between pixels and clustering centers. Experimental results on the brain images with different types and levels of noises demonstrate that the improved algorithm increases the accuracy of segmentation compared with the other soft clustering algorithms.
There will always exist a compromise between the network bandwidth and Web contents in the mobile Internet era. Web images need to be compressed efficiently by the servers and rendered fleetly by the clients. There ar...
详细信息
ISBN:
(纸本)9781538619377
There will always exist a compromise between the network bandwidth and Web contents in the mobile Internet era. Web images need to be compressed efficiently by the servers and rendered fleetly by the clients. There are several alternatives for the compression of Web images, from the long-existing coding standard JPEG to the emerging solution WebP. This paper presents a comparative analysis on both lossless compression and lossy compression of Web images based on the designed objective experiments, employing performance metrics including the Compression Ratio (CR), the Structural Similarity Index (SSIM), and the time consumption. The results show that WebP has competitive superiority over PNG in lossless image compression, however WebP and JPEG each has its own advantages in lossy compression.
With the development and improvement of imaging technology in the medical field, image technology, which provides important scientific basis for disease analysis, has become an indispensable part of disease diagnosis....
详细信息
ISBN:
(纸本)9781538619377
With the development and improvement of imaging technology in the medical field, image technology, which provides important scientific basis for disease analysis, has become an indispensable part of disease diagnosis. Therefore, how to dig out valuable information in these images and help doctors to make diagnosis more accurately and quickly have always been the concern of researchers. In this paper, we have made some improvements to the FCN network and incorporated Inception Architecture into it to build several convolutional neural networks. In our experiments, we trained the networks in IBSR dataset and contrasted the results with some classical methods. The results demonstrate that our improved network has high efficiency and accuracy in segmentation of MRI brain images.
In order to resolve the problem of poor real-time performance in the algorithm based on Luminance-contrast Transfer technique, the paper proposed a new image fusion method, which mainly uses lifting wavelet substituti...
详细信息
ISBN:
(纸本)9781538619377
In order to resolve the problem of poor real-time performance in the algorithm based on Luminance-contrast Transfer technique, the paper proposed a new image fusion method, which mainly uses lifting wavelet substituting biorthogonal wavelet. Moreover, by combining surreal video fusion with infrared and visible video, we propose a new algorithm based on Surreal Luminance-contrast Transfer technique. The algorithm extracts the background of the scene, and utilizes the background and the visible video sequence for surreal integration. Then we use the brightness contrast algorithm based on lifting wavelet and infrared video for the second fusion. Experiments show that the algorithm has good real-time performance, and works well in the smoky environment.
An improved image denoising algorithm based on block-matching and 3D collaborative filtering (BM3D) is proposed in this manuscript. Instead of using the same filtering model for all patches in an image, we employ two ...
详细信息
ISBN:
(纸本)9781538619377
An improved image denoising algorithm based on block-matching and 3D collaborative filtering (BM3D) is proposed in this manuscript. Instead of using the same filtering model for all patches in an image, we employ two different nonlocal filtering models in edge and smooth regions, respectively. We realize it by using the nonlocal centralized sparse representation (NCSR) to capture both local sparsity of wavelet coefficients and nonlocal similarity of grouped blocks. Experimental results demonstrate that the proposed method outperforms several state-of-the-art denoising methods in terms of objective metrics and visual quality.
We present a new algorithm for peak detection in ballistocardiographic (BCG) signals and use it for heart rate estimation. Systolic complexes of the BCG signal are enhanced and coarse heart beat locations estimated. E...
详细信息
We present a new algorithm for peak detection in ballistocardiographic (BCG) signals and use it for heart rate estimation. Systolic complexes of the BCG signal are enhanced and coarse heart beat locations estimated. Ejection waves I, J and K are detected simultaneously around coarse locations, only using detection of local maxima and weighted summation of peak heights. Due to a lack of reference BCG annotations, the algorithm's performance is assessed by using the detected peaks for heart rate estimation. On a dataset acquired with a pneumatic BCG system, we evaluate the heart rate estimation performance and compare the introduced algorithm against other methods found in literature. The dataset is gathered from 42 patients in a clinical environment and provides low-quality signals taken from a realistic scenario. With a mean absolute percentage error of 2.58 % at 65 % coverage, the presented method is on par with the best-performing state-of-the-art algorithms investigated. Limits of agreement (5th/95th percentiles) in a comparison with ECG-based heart rate measurements lie within P 5 = -3.63 and P 95 = 5.78 beat/min.
暂无评论