Showing papers by "Jeng-Shyang Pan published in 2010"
TL;DR: Two rotation invariant watermark embedding schemes in the non-subsampled contourlet transform (NSCT) domain based on the scale-adapted local regions are presented and can efficiently resist both signal processing attacks and geometric attacks.
44 citations
42 citations
17 Sep 2010
TL;DR: A fast algorithm to reduce the computation time of the temporal median operation by utilizing the characteristics of high correlation of adjacent frames, and a simple mechanism to check whether the median of the current frame is equal to that of the previous frame is designed.
Abstract: Temporal median filter is one of most popular background subtraction methods. However, median operation is very time-consuming which limits its applications. This paper presents a fast algorithm to reduce the computation time of the temporal median operation. By utilizing the characteristics of high correlation of adjacent frames, the fast algorithm designs a simple mechanism to check whether the median of the current frame is equal to that of the previous frame. The proposed algorithm reduces the computing frequency of median operations significantly, and the experimental results indicate it is much faster than the existing algorithms.
30 citations
13 Dec 2010
TL;DR: The experimental result indicates that the proposed FMO presents higher accuracy and convergence speed than the currently proposed PSO.
Abstract: In this paper, we propose Fish Migration Optimization (FMO) method based on a verified equation of the fish swim in the fish biology for solving numerical optimization problems. Inspired by the fish migration, the migration and the swim model are integrated into the optimization process. Four benchmark functions are used to test the convergence, the accuracy and the speed of FMO, and the experimental results are compared with PSO. The experimental result indicates that the proposed FMO presents higher accuracy and convergence speed.
29 citations
13 Dec 2010
TL;DR: A communication strategy for the parallelized Artificial Bee Colony (ABC) optimization is proposed for solving numerical optimization problems and increases the accuracy of the ABC on finding the near best solution.
Abstract: In this paper, a communication strategy for the parallelized Artificial Bee Colony (ABC) optimization is proposed for solving numerical optimization problems. The artificial agents are split into several independent subpopulations based on the original structure of the ABC, and the proposed communication strategy provides the information flow for the agents to communicate in different subpopulations. Three benchmark functions are used to test the behavior of convergence, the accuracy, and the speed of the proposed method. According to the experimental result, the proposed communicational strategy increases the accuracy of the ABC on finding the near best solution.
22 citations
10 Nov 2010
TL;DR: The experimental results demonstrate that the proposed ePSO effectively and fast tackles multi-objective optimization problem and shows the feasibility in real world.
Abstract: Acoustic communication networks in underwater environment are the key technology to explore global ocean. There are major challenges including (1) lack of stable and sufficient power supply, (2) disable of radio frequency signal and (3) no communication protocol designed for underwater environment. Thus, acoustic so far is the only media suitable to operate for underwater communication. In this paper, we study the technology of underwater acoustic communication to support underwater sensor networks. Toward the energy-effective goal, a cluster-based sensor network is assumed. The energy-dissipation of sensor nodes is optimized by biological computing such as Particle Swarm Optimization (PSO). The objective function of sensor node clustering is formulized to constraint on the network coverage and energy dissipation. The problem of dual-objective optimization is solved by the proposed extensible PSO (ePSO). ePSOis an innovation from traditional PSO. The major innovation is to offer an extensible particle structure and to enable more flexible search for optimal solutions in space. The experimental results demonstrate that the proposed ePSO effectively and fast tackles multi-objective optimization problem. The application of ePSO on underwater acoustic communication systems shows the feasibility in real world.
14 citations
TL;DR: A fast k-nearest neighbor search algorithm based on the wavelet transform, which exploits the important information hiding in the transform coefficients to reduce the computational complexity.
13 citations
TL;DR: This paper proposes a algorithm on vector quantization (VQ) based image watermarking, which is suitable for error-resilient transmission and can effectively overcome channel impairments while retaining the capability for ownership protection.
Abstract: W atermarking is one useful solution for digital rights management (DRM) systems , and it is a popular research topic in the last decade . In this paper, besides the inherent behavior to conquer against intentional or unintentional attacks for watermarking, we not only watch the survivability of embedded watermark, but also focus on retaining the watermarked image quality with the aid of multiple description coding (MDC). MDC is a technique for error resilient coding, suitable for transmitting compressed data over multiple channels. In this paper, we propose a n ew algorithm on vector quantization (VQ) based image watermarking, which is suitable for error-resilient transmission. By incorporating watermarking with MDC , the scheme we proposed for embedding three watermarks can effectively overcome channel impairments while retaining the capability for ownership protection. With the promising simulation results presented, we can demonstrate the utility and practicability of our algorithm.
10 citations
26 Sep 2010
TL;DR: An ant colony optimization- (ACO-) based routing algorithm to reduce power consumption is presented and the simulations indicate that the proposed algorithm obtains more balanced transmission among the nodes and reduces the power consumption of the routing.
Abstract: Power consumption is one of the most important problems for wireless sensor networks because of the battery limitation in each sensor. This paper presents an ant colony optimization- (ACO-) based routing algorithm to reduce power consumption. First, a grade table is built and referred to generate several possible routing paths. Then, the ACO explores these paths to reduce the power consumption of the nodes. The simulations indicate that the proposed algorithm obtains more balanced transmission among the nodes and reduces the power consumption of the routing.
9 citations
22 Nov 2010
TL;DR: A new approach based on XMPP and OSGi technology to home automation on Web is proposed and an application of fault-detection mechanism for smart refrigerators is demonstrated to illustrate the realization of the proposed approach.
Abstract: As the fast development of information and Internet technologies, lots of electrical appliances started to be digitized and combined with network technologies to provide people with much more multiple services. In this study, a new approach based on XMPP and OSGi technology to home automation on Web is proposed. An application of fault-detection mechanism for smart refrigerators is demonstrated to illustrate the realization of the proposed approach. Based on the proposed approach, a communication scheme of smart appliances is developed first. Then an inference engine is developed to derive the fault of refrigerator from the status of sensors. The appliance manufactures will be benefited from the fast self-diagnosis to remotely respond. The experimental results show the feasibility and modeling techniques applicable for the management of smart appliances on web.
8 citations
15 Oct 2010
TL;DR: A novel multibit image watermarking scheme resistant to rotation and scaling attacks is presented, using the Radon transform to correct the orientation of an image and embed and extract the watermark message in the DCT domain of the corrected image.
Abstract: In this paper, a novel multibit image watermarking scheme resistant to rotation and scaling attacks is presented. The Radon transform is used to correct the orientation of an image. Then the spread spectrum-based watermarking scheme which is scaling invariant is used to embed and extract the watermark message in the DCT domain of the corrected image. Experimental results show that the scheme possesses good robustness against rotation, scaling attacks and considerable robustness against typical image processing.
20 Aug 2010
TL;DR: In this article, the optimal economic dispatch (ED) through Cat Swarm Optimization (CSO) algorithm is studied for wind generator capacity integration into an isolated power system, and the CSO is also extended to coordinate wind and thermal dispatch and to minimize total generation cost.
Abstract: The aim of this research is to study the optimal economic dispatch (ED) through Cat Swarm Optimization (CSO) algorithm. Several techniques are applied to determine the proportion of wind generator capacity that can be integrated into an isolated system. The CSO is also extended to coordinate wind and thermal dispatch and to minimize total generation cost. Numerical experiments are included to demonstrate various constraints in production cost analysis and to provide valuable information for both operational and planning problems in the Taiwan power system.
13 Dec 2010
TL;DR: The theory and methods proposed in the past decade and their characteristics based on relevant literature are summarized and classified as quality indicator based approaches and statistic test approach here.
Abstract: Various randomized search heuristics have been proposed for multiobjective optimization problems. We need evaluate and compare the performance of these optimizers in order to make good use of them. This paper reviews the theory and methods proposed in the past decade and summarize their characteristics based on relevant literature. However, we didn’t list and analyze many methods proposed before because the relevant literature has done it. We look at them from the classification perspective. These assessment methods are classified as quality indicator based approaches and statistic test approach here. For quality indicators, we further classify them Pareto dominance compliant and non-compliant, unary and binary parameters. This enables us to choose suitable assessment measures in practice.
26 Sep 2010
TL;DR: A novel feature extraction algorithm, named two-dimensional exponential discriminant analysis (2DEDA), is proposed in this paper, which has higher recognition rate and lower computational complexity than the EDA.
Abstract: A novel feature extraction algorithm, named two-dimensional exponential discriminant analysis (2DEDA), is proposed in this paper. The 2DEDA is a generalization of exponential discriminant analysis (EDA). The 2DEDA is base on image matrices. So compared with the EDA, the 2DEDA has higher recognition rate and lower computational complexity. Experimental results demonstrate the advantages of 2DEDA.
10 Nov 2010
TL;DR: An integrated approach featuring automatic thresholding is developed and presented, and the experimental results indicate that the proposed method greatly improves the visual perceptibility as compared with previous approaches.
Abstract: Due to the inherent low-contrast in Electronic Portal Images (EPI), the perception quality of EPI has certain gap to the expectation of most physicians. It is essential to have effective post-processing methods to enhance the visual quality of EPI. However, only limited efforts had been paid to this issue in the past decade. To this problem, an integrated approach featuring automatic thresholding is developed and presented in this article. Firstly, Gray-Level Grouping (GLG) is applied to improve the global contrast of the whole image. Secondly, Adaptive Image Contrast Enhancement (AICE) is used to refine the local contrast within a neighborhood. Finally, a simple spatial filter is employed to reduce noises. The experimental results indicate that the proposed method greatly improves the visual perceptibility as compared with previous approaches.
20 Jun 2010
TL;DR: The statistical results from simulating data indicate the validity of modeling approach and the effects of three parameters on overhead and efficiency of system running come out in compared analysis both of different size swarms and of specific size swarm.
Abstract: As one of valid modeling and coordinated controlling tools, the particle swarm optimization algorithm can be extended for applying to task of swarm robotic search. To gain an insight into effects of key parameters on distributed swarm search, a series of simulations are conducted. In such control algorithm, main parameters including communication range, detection radius, and swarm size are paid close attentions. Similar to ideal “particles”, swarm robots are modeled at an abstract level with the extended particle swarm optimization method. Also, individual robots are designed to be controlled under three-state finite state machine mechanism having SingleSearch, SwarmSearch, and Declaration states. Then, control strategy and algorithm are developed. The statistical results from simulating data indicate the validity of modeling approach. Further, the effects of three parameters on overhead and efficiency of system running come out in compared analysis both of different size swarms and of specific size swarm.
01 Jan 2010
14 Jan 2010
TL;DR: Single directional two dimensional principal component analysis can extract the directional feature of face images more efficiently, so it gets a higher recognition rate, and experimental results demonstrate that the SD2DPCA and MD2D PCA have their advantages.
Abstract: In this paper, two novel face recognition frames are proposed, called single directional two dimensional principal component analysis (SD2DPCA) and multi-directional two dimensional principal component analysis (MD2DPCA). Compared with other popular algorithms, SD2DPCA needs less running time while achieves almost the same correct recognition rate. MD2DPCA can extract the directional feature of face images more efficiently, so it gets a higher recognition rate, and experimental results demonstrate that the SD2DPCA and MD2DPCA have their advantages.
10 Nov 2010
TL;DR: A robust image watermarking scheme by the use of k-means clustering, scale-invariant feature transform (SIFT) which is invariant to rotation, scaling, translation, partial affine distortion and addition of noise.
Abstract: In the traditional feature-base robust image watermarking, all bits of watermark message are bound with the feature point. If a few of points are attacked badly or lost, the performance of the watermarking scheme will decline or fail. In this paper, we present a robust image watermarking scheme by the use of k-means clustering, scale-invariant feature transform (SIFT) which is invariant to rotation, scaling, translation, partial affine distortion and addition of noise. SIFT features are clustered into clusters by k-means clustering. Watermark message is embedded bit by bit in each cluster. Because one cluster contains only one watermark bit but one cluster contains many feature points, the robustness of watermarking is not lean upon individual feature point. We use twice voting strategy to keep the robustness of watermarking in watermark detecting process. Experimental results show that the scheme is robust against various geometric transformation and common image processing operations, including scaling, rotation, affine transforms, cropping, JPEG compression, image filtering, and so on.
01 Dec 2010
TL;DR: Data Mining is a series of processes, which analyses the data and sieves some useful information or interesting knowledge out from real-world large and complex data sets, which results in that precisely extracting the knowledge or finding the relationships and patterns become more difficult.
Abstract: Data Mining (DM) is a series of processes, which analyses the data and sieves some useful information or interesting knowledge out from real-world large and complex data sets (Ghosh & Jain, 2005). Various statistics, analysis, and modeling methods are employed to find patterns and relationships in DM. The process of Knowledge Discovery from Data (KDD) is the key, which makes the outcome of DM being meaningful. Nevertheless, the fast development of computer science, the database management system (DBMS) and Data Warehouse (DW) pushes the size of the datasets increases forward with an astounding speed. It results in that precisely extracting the knowledge or finding the relationships and patterns become more difficult. Hence, the need for powerful tools to assist DM is clear. To build such ABSTRACT
19 Jul 2010
TL;DR: A novel robust image watermarking scheme is presented to embed a high capacity watermark into the feature point based characteristic regions and it can efficiently resist traditional signal processing attacks and geometric attacks.
Abstract: A novel robust image watermarking scheme is presented to embed a high capacity watermark into the feature point based characteristic regions. The watermark embedding positions are first determined by the scale-invariant feature transform (SIFT) based local circular regions. Then the binary watermark image is embedded by quantization in the Non-subsampled Contourlet Transform (NSCT) domain. In order to achieve rotation invariance, the watermark is embedded adaptively to the orientation of the region. Simulation results show that the proposed scheme can achieve high invisibility and it can efficiently resist traditional signal processing attacks and geometric attacks.
14 Jan 2010
TL;DR: Experimental results on ORL, YALE and UMIST face databases show that NKDA outperforms NDA on recognition, which demonstrates that it is feasible to improve NDA with kernel trick for feature extraction.
Abstract: Dimensionality reduction is the most popular method for feature extraction and recognition Recently, Li et al (IEEE PAMI, 2009) proposed Nonparametric Discriminant Analysis (NDA) based dimensionality reduction for face recognition and reported an excellent recognition performance However, NDA has its limitations on extracting the nonlinear features of face images for recognition, and owing to the highly nonlinear and complex distribution of face images under a perceivable variation in viewpoint, illumination or facial expression In order to increase the NDA, we extend the NDA with kernel trick to propose Nonparametric Kernel Discriminant Analysis (NKDA) for feature extraction and recognition Experimental results on ORL, YALE and UMIST face databases show that NKDA outperforms NDA on recognition, which demonstrates that it is feasible to improve NDA with kernel trick for feature extraction
01 Jan 2010
TL;DR: A framework for 3D fragmented object patching consisting of 3D shape feature extraction, 3D surface region segmentation and3D surface matching is proposed and the results show that the proposed algorithm is feasible and effective.
Abstract: Computer aided digital patching of the fragmented object has the great ad-
vantages in efficiency, re-operation and avoiding of inadvertent damage, which is widely
used in restoring and repairing of culture heritages. In this paper, we propose a uni�ed
framework for 3D fragmented object patching, and the contributions lies in: 1) a uni�ed
framework for 3D fragmented object patching consisting of 3D shape feature extraction,
3D surface region segmentation and 3D surface matching is proposed; 2) a novel geom-
etry projection based 3D histogram model is proposed to extract the shape feature of 3D
fragmented object robust to noise and sampling of 3D model; 3) a surface segmentation
based on region dilation method is presented with the enough considering of the in
uence
of surface coarseness on 3D surface region segmentation instead of handling the debris
with regular shape,
at surface and few broken surfaces using the current algorithms; 4)
a 3D surface matching based on height-map using 3D shape features directly instead of
using curves of debris as match features as the current algorithms. The experiments are
implemented on the simulation data and the real 3D scanning data of the fragmented
object with a Roland LPX-250 3D laser scanner, and the results show that the proposed
algorithm is feasible and effective.
13 Dec 2010
TL;DR: This paper shows a novel and low complexity approach for corner detection which is based on a normal vector of boundary fitting line and it avoids wrong detection of superfluous corners on no-corner arcs.
Abstract: this paper shows a novel and low complexity approach for corner detection which is based on a normal vector of boundary fitting line. It avoids wrong detection of superfluous corners on no-corner arcs. Our proposed method is superior to Sun’s k-cosine corner detection in detection time and has a better performance in localization. Our experiment results confirmed that the proposed approach of corner detection has reached our goal. It is free from rotation and able to locate the corner correctly. In addition, it also performs well for scaling images with the adjustable thresholds.
16 Apr 2010
TL;DR: Comparison results showed that particle swarm optimization with feasibility-based rules can get same optimal results as enumeration algorithm in much less calculation times.
Abstract: Crank block steering mechanism optimization is a nonlinear constrained optimization problem, which is important for forklift truck to get preferable steering performance. Particle swarm optimization (PSO) with feasibility-based rules is a swarm intelligent algorithm proposed to solve constrained optimization problems simply and effectively. Therefore, it is used to optimize crank block steering mechanism of forklift truck, which will try to minimize the maximal error of outer wheel steering angle, maximize the minimal transmission angle and improve the force transmission. Experimental results obtained by PSO with feasibility-based rules are compared with those obtained by enumeration algorithm which is reliable for optimization problems. The comparison results showed that particle swarm optimization with feasibility-based rules can get same optimal results as enumeration algorithm in much less calculation times.
10 Nov 2010
TL;DR: A novel embedded coding algorithm based on the reconstructed DCT coefficient to avoid the difficulties brought by the choice of wavelet transform base is proposed in this paper and its efficiency can be seen from the experimental results.
Abstract: As an efficient tool for image compression, wavelet has been widely used in all kinds of image processing areas. Based on the different encoding effects, wavelet compression algorithms can be probably classified into two categories. They are the embedded wavelet coding algorithms and the nonembedded wavelet coding algorithms. For the convenience of producing the anytime cut coding stream and the progressing reconstruction results, the embedded wavelet coding algorithms have been paid more attention in practice. Such as the embedded wavelet coding algorithms, EZW and SPIHT are the outstanding representatives. The only drawback for this wavelet based embedded coding algorithms is the choice of the different wavelet transform base. We propose a novel embedded coding algorithm based on the reconstructed DCT coefficient to avoid the difficulties brought by the choice of wavelet transform base in this paper. The new algorithm's efficiency can be seen from the experimental results.
01 Jan 2010
10 Nov 2010
TL;DR: A more robust scheme of audio watermarking which is based on cepstrum (or cepstral coefficients) and HOS (higher-order statistics) schemes is proposed, which could outperform the previous innovative one.
Abstract: In this paper, we propose a more robust scheme of audio watermarking which is based on cepstrum (or cepstral coefficients) and HOS (higher-order statistics) schemes. This scheme is a zero-watermarking one for the reason to maintain the audio quality. The audio signal is firstly kurtosis-estimated and feature-recognized, and then analyzed via CC and HOS, respectively, to extract the essential parameters and characteristics, which are then used for information embedding and extracting. The achievement of the proposed scheme could outperform the previous innovative one [1].
10 Nov 2010
TL;DR: A flexible MVC structure which can satisfy different bit rate requirements and an adaptive SI selection mechanism which better utilizes temporal and interview correlations to obtain more accurate SI are proposed.
Abstract: A GOP-flexible multiview video coding (MVC) scheme based on Wyner-Ziv (WZ) coding with adaptive side information (SI) is proposed in this paper. In this scheme, each view is WZ encoded independently at the encoder side, while the views are jointly decoded at the decoder side. Therefore, the communications of different cameras can be avoided at the encoder side. In WZ coding, SPIHT and LDPC are applied to improve compression efficiency. Meanwhile, a flexible MVC structure which can satisfy different bit rate requirements and an adaptive SI selection mechanism which better utilizes temporal and interview correlations to obtain more accurate SI are proposed. The experimental results show better rate-distortion performance of the proposed scheme than other tested schemes.
17 Sep 2010
TL;DR: A compatible scheme which compresses stereo video with flexible prediction structure and is compatible with the present equipments which are based on H.264 and process traditional monoview video.
Abstract: Stereo video can bring people into more vivid and actual sense, but the video data is too huge to storage and transmit conveniently. In this paper, a compatible scheme which compresses stereo video with flexible prediction structure is proposed. In the proposed scheme, we define a group of picture (GOP) considering of the real time of the video. There are different prediction structures according to temporal correlation and interview correlation whether the length of GOP is fixed. In any possible prediction structures, temporal correlations and interview correlations are utilized effectively. H.264 which is the most advanced video compression standard at present is utilized to improve the coding efficiency. The proposed scheme is compatible with the present equipments which are based on H.264 and process traditional monoview video. Therefore, the present equipments can process stereo video sequences conveniently. At last, the efficiency of the proposed scheme is proved by the experiment