摘要 :
The sufficient condition that guarantees perfect segmentation for an image with PCNNs when the intensity ranges of adjacent regions overlap is one of the main results presented in reference[1]. However, with deep understanding of ...
展开
The sufficient condition that guarantees perfect segmentation for an image with PCNNs when the intensity ranges of adjacent regions overlap is one of the main results presented in reference[1]. However, with deep understanding of the derivation process used in [1], it is shown in this paper that it is not a sufficient condition. The conditions for perfect image segmentation when there is an overlap in intensity ranges of adjacent regions are still remained unsolved.
收起
摘要 :
The sufficient condition that guarantees perfect segmentation for an image with PCNNs when the intensity ranges of adjacent regions overlap is one of the main results presented in reference[1]. However, with deep understanding of ...
展开
The sufficient condition that guarantees perfect segmentation for an image with PCNNs when the intensity ranges of adjacent regions overlap is one of the main results presented in reference[1]. However, with deep understanding of the derivation process used in [1]. it is shown in this paper that it is not a sufficient condition. The conditions for perfect image segmentation when there is an overlap in intensity ranges of adjacent regions are still remained unsolved.
收起
摘要 :
The Ising model is a classical model and it has been used in a number of problem domains, such as statistical physics and computer vision. The minimum energy of the Ising model is useful, however the lowest-energy is difficult to ...
展开
The Ising model is a classical model and it has been used in a number of problem domains, such as statistical physics and computer vision. The minimum energy of the Ising model is useful, however the lowest-energy is difficult to solve for the reason of time. In this paper, we use a polynomial-time algorithm based on the Ising model, to obtain the lowest energy and segment images efficiently. Image segmentation is a complex problem. There are many methods to solve this problem. It is the process of assigning a label to every pixel in an image such that pixels with the same label share certain visual characteristics. We will use the state of the Ising model to denote the image segmentation. The time complexity of this algorithm for image segmentation is polynomial-time.
收起
摘要 :
The Ising model is a classical model and it has been used in a number of problem domains, such as statistical physics and computer vision. The minimum energy of the Ising model is useful, however the lowest-energy is difficult to ...
展开
The Ising model is a classical model and it has been used in a number of problem domains, such as statistical physics and computer vision. The minimum energy of the Ising model is useful, however the lowest-energy is difficult to solve for the reason of time. In this paper, we use a polynomial-time algorithm based on the Ising model, to obtain the lowest energy and segment images efficiently. Image segmentation is a complex problem. There are many methods to solve this problem. It is the process of assigning a label to every pixel in an image such that pixels with the same label share certain visual characteristics. We will use the state of the Ising model to denote the image segmentation. The time complexity of this algorithm for image segmentation is polynomial-time.
收起
摘要 :
Wavelet transform is not applicable to arbitrarily-shaped region (or object) in images, due to the nature of its global decomposition. In this paper, the arbitrarily-shaped wavelet transform (ASWT) is proposed in order to solve th...
展开
Wavelet transform is not applicable to arbitrarily-shaped region (or object) in images, due to the nature of its global decomposition. In this paper, the arbitrarily-shaped wavelet transform (ASWT) is proposed in order to solve this problem and its properties are investigated. Computational complexity of the ASWT is also examined and it is shown that the ASWT requires significantly fewer computations than conventional wavelet transform, since the ASWT processes only the object region in the original image. Experimental results show that any arbitrarily-shaped image segment can be decomposed using the ASWT and perfectly reconstructed using the inverse ASWT.
收起
摘要 :
Wavelet transform is not applicable to arbitrarily-shaped region (or object) in images, due to the nature of its global decomposition. In this paper, the arbitrarily-shaped wavelet transform (ASWT) is proposed in order to solve th...
展开
Wavelet transform is not applicable to arbitrarily-shaped region (or object) in images, due to the nature of its global decomposition. In this paper, the arbitrarily-shaped wavelet transform (ASWT) is proposed in order to solve this problem and its properties are investigated. Computational complexity of the ASWT is also examined and it is shown that the ASWT requires significantly fewer computations than conventional wavelet transform, since the ASWT processes only the object region in the original image. Experimental results show that any arbitrarily-shaped image segment can be decomposed using the ASWT and perfectly reconstructed using the inverse ASWT.
收起
摘要 :
In a numerous image applications, image resizing is becoming indispensable as it impacts the bandwidth and storage requirements tremendously. The image resizing process introduces (or eliminates) many pixels to (or from) the origi...
展开
In a numerous image applications, image resizing is becoming indispensable as it impacts the bandwidth and storage requirements tremendously. The image resizing process introduces (or eliminates) many pixels to (or from) the original image. The values of such pixels and their positions must be carefully determined otherwise visible distortion and significant degradation in the quality of the image may occur. Several methods have been employed to resize images including pixel replication; linear interpolation; higher order interpolations, Bezier methods, DCT-based, wavelets and others. We introduce a new perceptually perfect image-resizing scheme that near optimally preserves edges and highly maintains the quality of homogenous regions. In this technique, the image is segmented via an efficient edge detector to produce an edge image and independent homogenous regions. The edge image is resized separately from the homogenous regions via chain coding and elaborate look-ahead-and-back tables technique. Homogenous regions are resized using a merciful adaptive region-based interpolation that exploits the characteristics of each region. At the end, the two parts are summed up to produce the desired resized image. Simulation results of numerous test images show that the proposed technique is subjectively and objectively far better than published results.
收起
摘要 :
In a numerous image applications, image resizing is becoming indispensable as it impacts the bandwidth and storage requirements tremendously. The image resizing process introduces (or eliminates) many pixels to (or from) the origi...
展开
In a numerous image applications, image resizing is becoming indispensable as it impacts the bandwidth and storage requirements tremendously. The image resizing process introduces (or eliminates) many pixels to (or from) the original image. The values of such pixels and their positions must be carefully determined otherwise visible distortion and significant degradation in the quality of the image may occur. Several methods have been employed to resize images including pixel replication; linear interpolation; higher order interpolations, Bezier methods, DCT-based, wavelets and others. We introduce a new perceptually perfect image-resizing scheme that near optimally preserves edges and highly maintains the quality of homogenous regions. In this technique, the image is segmented via an efficient edge detector to produce an edge image and independent homogenous regions. The edge image is resized separately from the homogenous regions via chain coding and elaborate look-ahead-and-back tables technique. Homogenous regions are resized using a merciful adaptive region-based interpolation that exploits the characteristics of each region. At the end, the two parts are summed up to produce the desired resized image. Simulation results of numerous test images show that the proposed technique is subjectively and objectively far better than published results.
收起
摘要 :
The standard separable two-dimensional (2-D) wavelet transform (WT) has recently achieved a great success in image processing because it provides a sparse representation of smooth images. However, it fails to capture efficiently o...
展开
The standard separable two-dimensional (2-D) wavelet transform (WT) has recently achieved a great success in image processing because it provides a sparse representation of smooth images. However, it fails to capture efficiently one-dimensional (1-D) discontinuities, like edges and contours, that are anisotropic and characterized by geometrical regularity along different directions. In our previous work, we proposed a construction of critically sampled perfect reconstruction anisotropic transform with directional vanishing moments (DVM) imposed in the corresponding basis functions, called directionlets. Here, we show that the computational complexity of our transform is comparable to the complexity of the standard 2-D WT and substantially lower than the complexity of other similar approaches. We also present a zerotree-based image compression algorithm using directionlets that strongly outperforms the corresponding method based on the standard wavelets at low bit rates
收起