Degradation in visibility is often introduced to images captured in poor weather conditions, such as fog or haze. To overcome this problem, conventional approaches focus mainly on the enhancement of the overall image contrast. However, because of the unspecified light-source distribution or unsuitable mathematical constraints of the cost functions, it is often difficult to achieve quality results. In this paper, a fusion-based transmission estimation method is introduced to adaptively combine two different transmission models. Specifically, the new fusion weighting scheme and the atmospheric light computed from the Gaussian-based dark channel method improve the estimation of the locations of the light sources. To reduce the flickering effect introduced during the process of frame-based dehazing, a flicker-free module is formulated to alleviate the impacts. The systematic assessments show that this approach is capable of achieving superior defogging and dehazing performance, compared with superior defogging and dehazing performance, compared with the state-of-the-art methods, both quantitatively and qualitatively.

In this study, an improved Block Truncation Coding (BTC) image compression scheme, namely Near-Aperiodic Dot-Diffused BTC (NADDBTC), is described. Firstly, the existing regular structures for the generation of bitmap are completely modified for aperiodic compressed results. Moreover, an adaptive quantization levels selection strategy and two parameters Class Matrix (CM) and Diffused Matrix (DM) for image compression are developed and co-optimized. The improvements produce results of superior image quality. Furthermore, the adaptive quantization levels are introduced for balanced false contour, impulsive noise, and blocking artifact. Experimental results demonstrate that the proposed NADDBTC is capable of providing excellent image quality and visual perception, as well as processing efficiency, similar to DDBTC by exploiting the innate parallelism advantage of dot diffusion.

This paper presents a new feature descriptor for vehicle verification. The object detection scheme generates the vehicle hypothesis (candidate) that requires subsequent confirmation in the vehicle verification stage with specific feature descriptors. In the procedure of vehicle verification, an image descriptor is generated from the statistical parameter of the curvelet-transformed (CT) subbands. The marginal distribution of CT output is a heavy-tailed bell-shaped function, which can be approximated as Gaussian, Laplace, and generalized Gaussian distribution (GGD) with high accuracy. The maximum likelihood estimation (MLE) produces the distribution parameters of each CT subband for the generation of the image feature descriptor. The classifier then assigns a class label for the vehicle hypothesis based on this descriptor information. As documented in the experimental results, this feature descriptor is effective and outperforms the existing methods in the vehicle verification tasks.

Formerly, dimensionality reduction techniques are effective ways for extracting statistical significance of features from their original dimensions. However, the dimensionality reduction also induces an additional complexity burden which may encumber the real efficiency. In this paper, a technique is proposed for the reduction of the dimension of samples rather than the features in the former schemes, and it is able to additionally reduce the computational complexity of the applied systems during the reduction process. This method effectively reduces the redundancies of a sample, in particular for those objects which possess partially symmetric property, such as human face, pedestrian, and license plate. As demonstrated in the experiments, based upon the premises of faster speeds in training and detection by a factor of 4.06 and 1.24, respectively, similar accuracies to the ones without considering the proposed method are achieved. The performance verifies that the proposed technique can offer competitive practical values in pattern recognition related fields.

In this paper, a halftoning-based watermarking method is presented. This method enables high pixel-depth watermark embedding, while maintaining high image quality. This technique is capable of embedding watermarks with pixel depths up to 3 bits without causing prominent degradation to the image quality. To achieve high image quality, the parallel oriented high-efficient direct binary search (DBS) halftoning is selected to be integrated with the proposed orientation modulation (OM) method. The OM method utilizes different halftone texture orientations to carry different watermark data. In the decoder, the least-mean-square-trained filters are applied for feature extraction from watermarked images in the frequency domain, and the naïve Bayes classifier is used to analyze the extracted features and ultimately to decode the watermark data. Experimental results show that the DBS-based OM encoding method maintains a high degree of image quality and realizes the processing efficiency and robustness to be adapted in printing applications.

The Adaboosting has attracted attention for its efficient face-detection performance. However, in the training process, the large number of possible Haar-like features in a standard sub-window becomes time consuming, which makes specific environment feature adaptation extremely difficult. This letter presents a two-stage hybrid face detection scheme using Probability-based Face Mask Pre-Filtering (PFMPF) and the Pixel-Based Hierarchical-Feature Adaboosting (PBHFA) method to effectively solve the above-mentioned problems in cascade Adaboosting. The two stages both provide far less training time than that of the cascade Adaboosting and thus reduce the computation complexity in face-detection tasks. In particular, the proposed PFMPF can effectively filter out more than 85% nonface in an image and the remaining few face candidates are then secondly filtered with a single PBHF Adaboost strong classifier. Given a${\rm M}\times{\rm N}$sub-window, the number of possible PBH features is simplified down to a level less than${\rm M}\times{\rm N}$, which significantly reduces the length of the training period by a factor of 1500. Moreover, when the two-stage hybrid face detection scheme are employed for practical face-detection tasks, the complexity is still lower than that of the integral-image based approach in the traditional Adaboosting method. Experimental results obtained using the gray feret database show that the proposed two-stage hybrid face detection scheme is significantly more effective than Haar-like features.

This paper describes the field testing of an underwater acoustic (UWA) modem in three distinct underwater environments: in the 5–7 m depth waters off the Goleta Pier in California, in the shallow 2–5 m depth waters of Viapahu Lagoon of Moorea, French Polynesia, and in a coral bommies field in Viapahu Lagoon. The modem employs M-ary direct-sequence spread-spectrum (DSSS) signaling, with joint detection and channel estimation performed by the matching pursuit (MP) algorithm. Our programmable DSP implementation of MP is shown to effectively estimate sparse multipath in real time. The current design employs a 24 kHz carrier and 7.8 kHz bandwidth to achieve a 161 bps data rate. Field tests show bit error-rates (BERs) ≤ 10−2are achievable in varied underwater environments at up to 440 m range with multipath spreads on the order of 9 msec.

In this paper, we present two novel robust methods for embedding watermarks into dithered halftone images. The first method is named Paired Subimage Matching Ordered Dithering (PSMOD), of which the decoder is provided with a priori information of the original watermark, and the corresponding application is copyright protection. The other method, Blind Paired Subimage Matching Ordered Dithering (BPSMOD), does not require the knowledge of the original watermark, and the main application is secret communication. Both methods utilize the bit and sub-subimage interleaving preprocesses. The experiments show that both techniques are sufficiently robust to guard against the cropping, tampering, and print-and-scan degradation processes, in either B/W or color dithered images. Both techniques are also sufficiently flexible for various levels of embedded capacities. Furthermore, a novel progressive coding scheme is also presented in this paper for the efficient display of dithered images. After the preprocessing of bit-interleaving, this algorithm utilizes the characteristic of reordered image to determine the transmitting order and then progressively reconstructs the dithered image. Moreover, the dithered images are further compressed by lossy and lossless procedures. The experimental results demonstrate high-quality reconstructions while maintaining low transmitted bit rates.

Channel estimation and multiuser detection are enabling technologies for future generations of wireless applications. However, sophisticated algorithms are required for accurate channel estimation and multiuser detection, and real-time implementation of these algorithms is difficult. This paper presents architectural design methods for wireless channel estimation which can be leveraged to enable real-time multiuser detection. We redesign the matching pursuit (MP) channel estimation algorithm to reduce the complexity while maintaining the estimation accuracy. Furthermore, we develop a parameterized intellectual property (IP) core, which provides a hardware implementation of the MP algorithm. Experimental results demonstrate the effectiveness and ef- ficiency of the new algorithm and IP core for channel estimation. The implementation of our MP core on a modern, high performance reconfigurable system is about 216 times faster than running the algorithm on a state of the art microprocessor. The MP core possesses the speed required for performing true multiuser detection, enabling future generations of wireless communication applications.

A halftone watermarking technique of high capacity, robustness, and capacity flexibility is presented in this paper. This parity-matched error diffusion (PMEDF) method is capable of achieving an embedded capacity as high as 6.25% to 25% with good image quality and without the original image as the reference to decode the watermark. As the experimental results demonstrated, this technique is able to guard against degradation due to cropping, tampering, and printed-and-scanned process in error-diffused halftone images.

Channel estimation and distributed positioning algorithms are presented for geolocation in a wireless ad hoc network. The network uses a direct-sequence code-division multiple-access-based handshaking protocol, in which nodes receive multiple acknowledgment packets in response to a request-to-send waveform. Round-trip travel time (RTT) and angle-of-arrival (AOA) measurements are obtained using the generalized successive interference cancellation/matching pursuits (GSIC/MP) algorithm. The performance of GSIC/MP is evaluated via simulation and comparison to the Crame/spl acute/r-Rao bound. Position estimates are initialized using linearized least-squares and updated by an extended Kalman filter-based algorithm that includes measurement validation for nonline-of-sight error mitigation. The method is generalized for distributed estimation in sparsely connected networks: at each node, position estimates from connected nodes are incorporated via a fusion algorithm and updated using locally processed RTT/AOA measurements. Finally, comprehensive ad hoc network simulations are presented including channel ray tracing, RTT/AOA estimation and validation, and distributed positioning.

In this letter, we present a novel robust method for embedding watermarks into dithered halftone images. The method is named paired sub-image matching ordered dithering (PSMOD), of which the decoder is provided with a priori information of the original watermark. The method utilizes the bit and sub-subimage interleaving preprocesses. The experiments show that the technique is sufficiently robust to guard against the cropping, tampering, and printed-and-scanned degradation processes, in either B/W or color dithered images. This technique is also sufficiently flexible for various levels of embedded capacities.

Scheduling resources on Grids is a well-known problem. The extension of Grids, to LambdaGrids requires scheduling of lambdas, i.e., end-to-end high-speed circuits. In this paper we propose an heuristic for the scheduling of lambdas specifically for file transfers, given that many eScience applications require high-throughput transfers of large files. We call this heuristic varying-bandwidth list scheduling (VBLS) because the scheduler returns a time-range-capacity (TRC) allocation vector with varying banchwidth levels assigned for different time ranges within the duration of a transfer. The advantage of VBLS over a fixed-bandwidth allocation scheme is that it allows the scheduler to backfill any holes left in resource allocations. Such a scheme is enabled by end host applications specifying the file size in their transfer requests. To characterize VBLS, we used analytical models and ran simulations. Our results show that VBLS performance is close to packet-switching performance, which means that file transfers can take advantage of bandwidth that becomes available subsequent to the start of transfers, a critical drawback of typical fixed-bandwidth allocation schemes in circuit-switched networks.

An advantage of CDMA-based protocols is that packets can be transmitted immediately without prior channel access or resource reservation. In this paper, however, we propose to incorporate a reservation mechanism into CDMA-based protocols for next-generation wireless networks with data, voice, and multimedia applications. The resultant paradigm is called code division multiple access with adaptable reservation (CDMA/AR). The motivations for this change include that QoS guarantees can continue to be provisioned in CDMA/AR-based systems even when many handoff mobile stations enter an already heavily utilized cell. Our simulation results and mathematical analysis show that the adaptable reservation and differentiated admission control mechanisms of CDMA/AR can considerably reduce the handoff-rejection and call-blocking rates for higher-priority traffic to very small values, while achieving throughput close to 100% at the same time.

Video-endoscopy, a mode of minimally invasive surgery, has proven to be significantly less invasive to the patient. However, it creates a much more complex operation environment that requires the surgeon to operate through a video interface. Visual feedback control and image interpretation can be difficult. Poor visual feedback in video-endoscopy prolongs the operation time, increases the risk to the patient, and drives up the cost of health care. It is a major roadblock in replacing the traditional, highly traumatic open surgical procedures with the much less invasive, more patient friendly video-endoscopy, and in training the surgeons to master this new made of operation. Our research objective is thus to design, code, and validate on real images novel image analysis and rectification algorithms to enhance the visual feedback to the surgeon in video-endoscopy.

We present a scheme for optimal VLSI layout and packaging of butterfly networks under the Thompson model, the multilayer grid model, and the hierarchical layout model. We show that when L layers of wires are available, an N-node butterfly network can be laid out with area 4N2/L2 log22 N + o (N2/L2 log2 N), maximum wire length 2N/L log2 N + o (N/L log N), and volume 4N2/L log22 N + o (N2/L log2 N) , under the multilayer 2-D grid model, where only one active layer (for network nodes) is required and L layers of wires are available.

The dynamic broadcast problem is the communication problem where source packets to be broadcast to all the other nodes are generated at each node of a parallel computer according to a certain random process, such as a Poisson process. The lower bounds on the average reception delay required by any oblivious dynamic broadcast algorithm in a d-dimensional hypercube are /spl Omega/(d+1/1-/spl rho/) when packets are generated according to a Poisson process, where p is the load factor. The best previous algorithms for hypercubes only achieve /spl Omega/(d/1-/spl rho/) average reception delay. In this paper, we propose dynamic broadcast algorithms that require optimal O(d+1/1-/spl rho/) average reception delay in d-dimensional hypercubes and n/sub 1//spl times/n/sub 2//spl times//spl middot//spl middot//spl middot/n/sub d/ tori with n/sub i/=O(1). We apply the proposed broadcast scheme to a variety of other network topologies for efficient dynamic broadcast and present several methods for assigning priority classes to packets.

This paper describes the multiple frequency-band tomographic backward propagation technique for sonar imaging. This technique models the received array time data as the complex superposition of plane waves of different wavelengths. The received plane waves are individually backward propagated to the image space to form single frequency holographic reconstructions. The individual holograms are coherently superimposed to form a final tomographic reconstruction of the image with superior resolution over the single-frequency holographic images. This paper outlines the backward propagation technique for sonar imaging, derives a quantitative description of performance under wide bandwidth conditions, and highlights the similarities of this technique with the popular phase-shift beamforming algorithm.

The scanning tomographic acoustic microscope (STAM) was proposed in 1982 as a method of improving the resolution capability of the scanning laser acoustic microscope (SLAM) based on the principles of tomography. By modifying the SLAM with a quadrature detector, tomographic projections that contain both the amplitude and phase information of the scattered wavefield can be acquired. Subsequently, multiple projections acquired with different incident waves are combined using the "back-and-forth" propagation algorithm to form the tomographic reconstruction. The first STAM reconstructions have been obtained to experimentally demonstrate the superior resolution capability of the STAM over the SLAM. In this paper, the implementation of the STAM is described, and experimental reconstructions of multiple-layer specimens are demonstrated.

This paper presents a three-level cascade algorithm for estimating and compensating platform motion errors during synthetic aperture sonar imaging. By using a multiple element receiving array, physical aperture images can be produced from each transmit burst. With proper spatial sampling, there is sufficient redundancy in successive physical aperture images to extract motion parameters that can be applied to the image data to fully realize the improved resolution capability of synthetic aperture techniques. Since this algorithm is based on autofocus techniques, changes in the medium during synthetic aperture formation are inherently corrected as well. In addition, this algorithm is compared to two motion estimation and compensation techniques used in synthetic aperture radar imaging.>

The main problems involved in synthetic aperture sonar imaging are medium instability, slow mapping rate, and platform motion. Proposed solutions to these problems are reviewed, and a new technique for synthetic aperture sonar imaging is presented which allows a fast mapping rate and improved motion compensation. This technique involves using multiple receivers to form a complex-valued preliminary image for each pulse. The preliminary images are used to estimate translation and rotation, and then they are registered, phase corrected, and superimposed to form high-resolution synthetic aperture sonar images.>

Tomographic reconstruction of ionospheric distributions, such as electron densities, has generated considerable interest due to the possibility of reconstructing two-dimensional images from existing data. Previous remote sensing methods such as ionosonde, differential Doppler, and Faraday rotation have been capable only of one-dimensional reconstructions. An algorithm based upon the filtered backprojection algorithm of X-ray tomography has been developed that provides such two-dimensional images; however, the nature of the ionospheric system limits its resolving capability. The ionospheric tomography imaging system is compared to the traditional X-ray tomography system to determine its fundamental resolution degradation parameters. These parameters are then isolated in order to examine their individual effects on the reconstructed images.>

The scanning laser acoustic microscope (SLAM) has recently been equipped with a quadrature detector for holographic signal detection. It has been determined that data acquisition error often exists in the quadrature detector output. The resulting degradation in the reconstructed images is in the form of a fixed-frequency background fluctuation in the laser scar direction. The authors present an algorithm for the estimation and correction of the data acquisition error for the purpose of resolution improvement. The error is estimated by comparing the single-sideband wave-field signals of the quadrature output pair. The error correction is achieved by a matrix operator applied to the output pair. Examples involving experimental data are included to demonstrate the removal of the wave-field error and the improved resolution of the reconstructed images.>

The authors present a technique that improves the computational efficiency of an algorithm used for discrete bandlimited signal extrapolation. It consists of extrapolating segments or subsequences of the original data set and then uncorrelating the individual results to yield a final estimate of the signal. It is found that the method is provides high-quality reconstructions while reducing the amount of computation required. However, it results in reduced resolution in the final reconstruction. Therefore, a tradeoff exists between computational speed and resolution in the final estimate.>

The authors describe an algorithm which uses a Gaussian and mean curvature profile for extracting special points on terrain and then use these points for recognition of particular regions of the terrain. The Gaussian and mean curvatures are chosen because they are invariant under rotation and translation. In the Gaussian and mean curvature image, the points of maximum and minimum curvature are extracted and used for matching. The stability of the position of those points in the presence of noise and with resampling is investigated. The input for this algorithm consists of 3-D digital terrain data. Curvature values are calculated from the data by fitting a quadratic surface over a square window and calculating directional derivatives of this surface. A method of surface fitting which is invariant to coordinate system transformation is suggested and implemented. The algorithm is tested with and without the presence of noise, and its performance is described.>

An algorithm is presented which uses Gaussian curvature for extracting special points on the terrain, and then uses these points for recognition of particular regions of the terrain. The Gaussian curvature is chosen because it is invariant under isometry, which includes rotation and translation. In the Gaussian curvature image, the points of maximum and minimum curvature are extracted and used for matching. The stability of the position of these points in the presence of noise with resampling is investigated. The Gaussian curvature is calculated from the 3-D digital terrain data by fitting a quadratic surface over a square window and calculating directional derivatives of this surface. A method of surface fitting which is invariant to coordinate system transformation is suggested and implemented. This method involves finding an optimal directional in which the fitting is performed.>

A description is given of a curvature-based approach to motion analysis of nonrigid surfaces. Based on changes in the mean and Gaussian curvatures during the motion, it is possible to classify the motion of a surface at each point as rigid, isometric, homothetic, conformal, and general (nonconformal). The general theory of curvature changes for all types of motions is presented. Then the formula specifying changes in the Gaussian curvature during homothetic motion is derived. Also, two special cases of nonrigid motion are considered. The first case deals with piecewise rigid motion. An algorithm is presented which separates a piecewise rigid surface into its rigid parts. The second case deals with homothetic motion. A method is given to isolate regions of the surface where the homothetic assumption is violated. Both algorithms are tested on simulated data and results.>

The optimal deconvolution filter for high-resolution nuclear magnetic resonance (NMR) spectral parameter estimation is formulated. In the implementation of the algorithm, the two-dimensional parameter distribution reconstructed with the inverse filter is used as the probability density distribution for the formulation of the optimal filter. The deconvolution method for high-resolution NMR spectroscopy is first reviewed. Subsequently, the optimal filter process is modeled as an estimation procedure based on data over a finite-size observation aperture. The optimal filter improve the accuracy of the estimation of the complex amplitudes of the signal components. Simulated reconstructions were used to demonstrate the improvement of the resolving capability.>

Three algorithms for support-limited phase retrieval from Fourier magnitude are compared by computer simulation using small test arrays. It was found that Fienup's hybrid input-output algorithm is superior in terms of both computer time requirement and noise sensitivity.

In this paper, we present a motion estimation technique for coherent underwater acoustic imaging systems. This technique does not require point feature selection or matching correspondence identification. The rotation matrix is estimated from the covariance matrices of the amplitude distributions of the wavefield spectra, and the translation vector can be obtained subsequently from the phase change in the spatial-frequency domain.

In this paper, we provide an alternative viewpoint of NMR spectral parameter estimation that we describe the estimation process as a two-dimensional deconvolution operation. This technique gives superior resolution and requires relatively simple computation.

In this paper, we present a technique to improve the performance of discrete band-limited signal extrapolation by recursively changing the associate weighting function. During the process, the weighting is replaced iteratively by the amplitude of the spectral distribution of the previous estimate. This alters the sensitivity of the algorithm and results in considerable enhancement of the performance.

The discrete band-limited signal extrapolation is a finite-dimensional linear estimation operation. In this paper, we describe the algorithm as a two-step recognition and reconstruction process and analyze its performance and limitation by using the associated singular values as indicators of the algorithm's sensitivity and capability for signal identification. We also indicate that the performance of the algorithm depends on the dimension of the effective vector space, and provide the relationship between the variation of the dimension and control-point spacings.

Two generalized discrete band-limited signal extrapolation methods are presented. These methods are capable of altering the structure of the eigenvectors as well as the distribution of the eigenvalues of the linear matrix operation, and thus improving the extrapolation process.