With an annual growth of billions of sensor-based devices, it is an urgent need to do stream mining for the massive data streams produced by these devices. Cloud computing is a competitive choice for this, with powerful computational capabilities. However, it sacrifices real-time feature and energy efficiency. Application-specific integrated circuit (ASIC) is with high performance and efficiency, which is not cost-effective for diverse applications. The general-purpose microcontroller is of low performance. Therefore, it is a challenge to do stream mining on these low-cost devices with scalability and efficiency. In this paper, we introduce an FPGA-based scalable and parameterized architecture for stream mining.Particularly, Dynamic Time Warping (DTW) based k-Nearest Neighbor (kNN) is adopted in the architecture. Two processing element (PE) rings for DTW and kNN are designed to achieve parameterization and scalability with high performance. We implement the proposed architecture on an FPGA and perform a comprehensive performance evaluation. The experimental results indicate thatcompared to the multi-core CPU-based implementation, our approach demonstrates over one order of magnitude on speedup and three orders of magnitude on energy-efficiency.

A conventional HMM-based speech synthesis system for Hanoi Vietnamese often suffers from hoarse quality due to incomplete F0 parameterization of glottalized tones. Since estimating F0 from glottalized waveform is rather problematic for usual F0 extractors, we propose a pitch marking algorithm where pitch marks are propagated from regular regions of a speech signal to glottalized ones, from which complete F0 contours for the glottalized tones are derived. The proposed F0 parameterization scheme was confirmed to significantly reduce the hoarseness whilst slightly improving the tone naturalness of synthetic speech by both objective and listening tests. The pitch marking algorithm works as a refinement step based on the results of an F0 extractor. Therefore, the proposed scheme can be combined with any F0 extractor.

To analyze the structure of a set of high-dimensional perfect sequences over a composition algebra over R, we developed the theory of Fourier transforms of the set of such sequences. We define the discrete cosine transform and the discrete sine transform, and we show that there exists a relationship between these transforms and a convolution of sequences. By applying this property to a set of perfect sequences, we obtain a parameterization theorem. Using this theorem, we show the equivalence between the left perfectness and right perfectness of sequences. For sequences of real numbers, we obtain the parameterization without restrictions on the parameters.

The simple repetitive control system proposed by Yamada et al. is a type of servomechanism for periodic reference inputs. This system follows a periodic reference input with a small steady-state error, even if there is periodic disturbance or uncertainty in the plant. In addition, simple repetitive control systems ensure that transfer functions from the periodic reference input to the output and from the disturbance to the output have finite numbers of poles. Yamada et al. clarified the parameterization of all stabilizing simple repetitive controllers. Recently, Yamada et al. proposed the parameterization of all stabilizing two-degrees-of-freedom (TDOF) simple repetitive controllers that can specify the input-output characteristic and the disturbance attenuation characteristic separately. However, when using the method of Yamada et al., it is complex to specify the low-pass filter in the internal model for the periodic reference input that specifies the frequency characteristics. This paper extends the results of Yamada et al. and proposes the parameterization of all stabilizing TDOF simple repetitive controllers with specified frequency characteristics in which the low-pass filter can be specified beforehand.

A parameterization of perfect sequences over composition algebras over the real number field is presented. According to the proposed parameterization theorem, a perfect sequence can be represented as a sum of trigonometric functions and points on a unit sphere of the algebra. Because of the non-commutativity of the multiplication, there are two definitions of perfect sequences, but the equivalence of the definitions is easily shown using the theorem. A composition sequence of sequences is introduced. Despite the non-associativity, the proposed theorem reveals that the composition sequence from perfect sequences is perfect.

A perfect array is an array for which the autocorrelation function is impulsive. A parameterization of perfect arrays of real numbers is presented. Perfect arrays are represented by trigonometric functions. Three formulae are obtained according to the parities of the size of the array. Examples corresponding to each formula are shown. In the case of 66 arrays, the existence of a set of perfect arrays having integer components is shown.

A perfect sequence is a sequence having an impulsive autocorrelation function. Perfect sequences have several applications, such as CDMA, ultrasonic imaging, and position control. A parameterization of a perfect sequence is presented in the present paper. We treat a set of perfect sequences as a zero set of quadratic equations and prove a decomposition law of perfect sequences. The decomposition law reduces the problem of the parameterization of perfect sequences to the problem of the parameterization of quasi-perfect sequences and the parameterization of perfect sequences of short length. The parameterization of perfect sequences for simple cases and quasi-perfect sequences should be helpful in obtaining a parameterization of perfect sequences of arbitrary length. According to our theorem, perfect sequences can be represented by a sum of trigonometric functions.

In 3D computer graphics, mesh parameterization is a key technique for digital geometry processings such as morphing, shape blending, texture mapping, re-meshing and so on. Most of the previous approaches made use of an identical primitive domain to parameterize a mesh model. In recent works of mesh parameterization, more flexible and attractive methods that can create direct mappings between two meshes have been reported. These mappings are called "cross-parameterization" and typically preserve semantic feature correspondences between target meshes. This paper proposes a novel approach for parameterizing a mesh into another one directly. The main idea of our method is to combine a competitive learning and a least-square mesh techniques. It is enough to give some semantic feature correspondences between target meshes, even if they are in different shapes or in different poses.

A simple and an efficient algorithm for polygon morphing is proposed in this paper. We adopt the parametric curve representation based on Fourier parameter estimation to transfer the traditional morphing process in spatial domain to a process in the parametric space instead. The principles are to express the polygon as the union of matching segments that are described by the estimated Fourier parameters. We have also designed a data resampling method that effectively controls the shape morphing according to the corresponding curvature values. Intermediate objects in-between the source and target polygons are then constructed based on the interpolation of Fourier parameters of the two polygons. Fourier parameters of the resampled polygons can be obtained efficiently by using the fast Fourier transform (FFT) algorithm. The experimental results show that the appearances of the morphed objects are superior to the ones obtained by the methods available.

The warp model of the extended Hough transform (EHT) has been proposed to design the explicit expression of the transform function of EHT. The warp model is a skewed parameter space (R(µ,ξ), φ(µ,ξ)) of the space (µ,ξ), which is homeomorphic to the original (ρ,θ) parameter space. We note that the introduction of the skewness of the parameter space defines the angular and positional sensitivity characteristics required in the detection of lines from the pattern space. With the intent of contributing some solutions to basic computer vision problems, we present theoretically a dynamic and centralfine/peripheral-coarse camera vision architecture by means of this warp model of Hough transform. We call this camera vision architecture askant vision' from an analogy to the human askant glance. In this paper, an outline of the EHT is briefly shown by giving three functional conditions to ensure the homeomorphic relation between (µ,ξ) and (ρ,θ) parameter spaces. After an interpretation of the warp model is presented, a procedure to provide the transform function and a central-coarse/peripheralfine Hough transform function are introduced. Then in order to realize a dynamic control mechanism, it is proposed that shifting of the origin of the pattern space leads to sinusoidal modification of the Hough parameter space.