Convolution kernels for trees provide effective means for learning with treestructured data, such as parse trees of natural language sentences. Unfortunately, the computation time of tree kernels is quadratic in the size of the trees as all pairs of nodes need to be compared: large trees render convolution kernels inapplicable. In this paper, we propose a simple but efficient approximation tech...

In the wireless communication, the error detection and correction techniques are widely used to achieve higher transmission efficiency. While transmitting the data, some coding techniques are deployed that provides the coding gain depending on its error correcting capability. Coding gain allows the wireless communication at lower transmit power level while maintaining low BER. Reed Solomon – Co...

Convolution false-targets jamming against LFM fire-control radars generates range and velocity false targets which are coherent with target echo, which increases the difficulty of jamming detection and suppression and makes the victim radar system lost the track of real target. To combat against this type of jamming, the uncorrelated characteristic between the jamming and echo in FRFT domain is...

This paper presents a novel fully automated process of features extraction and classification of Multiple sclerosis (MS) daisies from magnetic resonance images (MRI). This hybrid method uses convolution neural network (CNN) for features extraction and a multilayer neural network for classification two classes normal and MS. The convolution neural network for recognition of Multiple sclerosis is...

This paper proposes two new methods to detect the fingerprints of different persons based on one-dimensional and two-dimensional discrete wavelet transformations (DWTs). Recent literature shows that fingerprint detection based on DWT requires less memory space compared to pattern recognition and moment-based image recognition techniques. In this study four statistical parameters cross correlati...

In a recent paper by Chen et al., the authors present a clever means of decomposing the Laplacian-of-Gaussian (LOG) kernel into the product of a Gaussian and a (smaller) LOG mask. They then proceed to develop a fast algorithm for convolution which exploits thespatial frequency properties of these operators to allow the image tobe decimated (subsampled). Although this approach is both no...

Face recognition is one of the most challenging tasks in the field of image processing. This paper presents an optimized edge detection algorithm for the task of face recognition. In this method a gradient based filter using a wide convolution kernel is applied on the image to extract the edges. Later a thinning algorithm optimized for the wide convolution kernel is applied on the extracted edg...

Partial Fourier reconstruction algorithms exploit the redundancy in magnetic resonance data sets so that half of the data is calculated during image reconstruction rather than acquired. The conjugate synthesis, Margosian, homodyne detection, Cuppen and POCS algorithms are evaluated using spatial frequency domain analysis to show their characteristics and where limitations may occur. The phase c...

This paper proposed an edge detection scheme which is deduced from Fresnel diffraction. Analysis in this paper shows that Fresnel convolution kernel function performs well on edge enhancement when images are transformed into complex functions. Due to its mathematical complexity, the method is simplified into a linear convolution filter. The new edge detector is designed based on the simplified ...