As global online access to information becomes more common, the technology ofmultilingual optical character recognition (OCR) increases in importance as a way to convert paperdocuments into electronic, searchable, tex...
详细信息
As global online access to information becomes more common, the technology ofmultilingual optical character recognition (OCR) increases in importance as a way to convert paperdocuments into electronic, searchable, text. In OCR, as in any evolving technology, carefulevaluation is an integral part of research and development. OCR evaluation is done by comparing asystem's output for a dataset of document test images with the corresponding correct symbolic text,known as ground truth. Unfortunately, the usual way of obtaining ground truth is by manual dataentry by humans, which is labor-intensive, time-consuming, expensive, and prone to errors. Worse,because no single set of ground truth evaluation data can be used in more than one language, therehas until now been no way to conduct carefully controlled OCR experiments in a multilingual setting.
An assignment-based solution for the data association problem in synchronous passive multisensor (Type 3) tracking systems involves two steps: first measurement-to-measurement or static association is solved using a m...
详细信息
An assignment-based solution for the data association problem in synchronous passive multisensor (Type 3) tracking systems involves two steps: first measurement-to-measurement or static association is solved using a multidimensional (S-dimensional or S-D with S sensors) assignment, and then measurement-to-track association is solved using a 2-D assignment. This solution is computationally very expensive and, to rectify an efficient (S + 1)-D assignment algorithm has been proposed in the literature. Two new assignment-based algorithms are proposed that use prior track information (i.e., predicted state and covariance) which result in improved tracking performance compared with the existing solutions, while requiring considerably less computations. One of the proposed algorithms, the gated assignment, is similar to the two-step solution mentioned above except that it uses prior track information and avoids the need to consider all possible association hypotheses in the static association step. The second algorithm, the gated (S + 1)-D assignment, combines the gated assignment and the (S + 1)-D algorithms. An approximation to the (S + 1)-D algorithm is also derived when sensor measurements are independent, which results in an extremely fast solution. Simulation results confirm that the proposed algorithms show improved tracking performance and faster execution times.
Data-accumulating algorithms (d-algorithms for short), extensively studied in [12], work on an input considered as a virtually endless stream. The computation terminates when all the currently arrived data have been p...
详细信息
Data-accumulating algorithms (d-algorithms for short), extensively studied in [12], work on an input considered as a virtually endless stream. The computation terminates when all the currently arrived data have been processed before another datum arrives. In this paper a finer characterization of the class of d-algorithms is given, and it is shown that this class is identical to the class of on-line algorithms under a proper definition of the latter. The parallel implementation of d-algorithms is then investigated. It is found that, in general, the speedup achieved through parallelism can be made arbitrarily large for almost any such algorithm. On the other hand, we prove that for d-algorithms whose static counterparts manifest only unitary speedup, no improvement is possible through parallel implementation.
The article presents a discussion of topics in computer science related to Boolean satisfiability. It focuses on solutions which can be readily used in applications software, and their effectiveness for solving certai...
详细信息
The article presents a discussion of topics in computer science related to Boolean satisfiability. It focuses on solutions which can be readily used in applications software, and their effectiveness for solving certain types of problems. Topics addressed include the interaction between necessary values and constraints in generating computational complexity. Examples are provided using decision trees solved using various computer algorithms. The increasing commercial use of such programs in fields such as computer hardware and software testing and development is noted.
Previous work on autonomous formation flying guidance and control identified three key challenges to overcome in order to obtain a fully autonomous guidance and control loop: an accurate but simple model of relative m...
详细信息
Previous work on autonomous formation flying guidance and control identified three key challenges to overcome in order to obtain a fully autonomous guidance and control loop: an accurate but simple model of relative motion about elliptical and perturbed orbits, an efficient way of performing conflicting requirements trade-off with power-limited on-board computers, and finally an optimal or near-optimal control algorithm easy to implement on a flight computer. This paper first summarizes recent developments on each of these subject that help to overcome these challenges, developments which are then used as building blocks for an autonomous formation flying guidance and control system. This system autonomously performs trade-offs between conflicting requirements. i.e. minimization of fuel cost, formation accuracy and equal repartition of the fuel expenditure within the formation. Simulation results show that a complete guidance and control loop can be established using mainly analytical results and with very few numerical optimization which facilitates on-board implementation. (c) 2008 Elsevier Ltd. All rights reserved.
Selective encryption is a technique to save computational complexity or enable interesting new system functionality by only encrypting a portion of a compressed bitstream while still achieving adequate security. Altho...
详细信息
Selective encryption is a technique to save computational complexity or enable interesting new system functionality by only encrypting a portion of a compressed bitstream while still achieving adequate security. Although suggested in a number of specific cases, selective encryption could be much more widely used in consumer electronic applications ranging from mobile multimedia terminals through digital cameras were it subjected to a more thorough security analysis. We describe selective encryption and develop a simple scalar quantizer example to demonstrate the power of the concept, list a number of potential consumer electronics applications, and then describe an appropriate method for developing and analyzing selective encryption for particular compression algorithms. We summarize results from application of this method to the MPEG-2 video compression algorithm.
This study addresses the trajectory tracking control problem of electrically driven wheeled mobile robots under non-holonomic constraints in the presence of model uncertainties without velocity measurement. By definin...
详细信息
This study addresses the trajectory tracking control problem of electrically driven wheeled mobile robots under non-holonomic constraints in the presence of model uncertainties without velocity measurement. By defining a suitable set of output equations, a new input-output model of wheeled mobile robots is developed, which helps the designer utilise the classic control algorithms of robot manipulators. An observer-based trajectory tracking controller is proposed for the new wheeled mobile robot (WMR) model. Then, in order to reduce the design complexity, the dynamic surface control approach is effectively exploited to propose a tracking controller considering the actuator dynamics. Adaptive robust techniques are also adopted to cope with the parametric and non-parametric uncertainties in the WMR model. A Lyapunov-based stability analysis is utilised to guarantee that tracking and state estimation errors are uniformly ultimately bounded. Simulation results are presented to illustrate the feasibility and efficiency of the proposed controller.
A quadratic-speed-dependent Voigt line shape (qSDV) with line mixing (qSDV+LM), together with spectroscopic line parameters from Devi et al. [1,2] for the 2v(3) band of CH4, was used to retrieve total columns of CH4 f...
详细信息
A quadratic-speed-dependent Voigt line shape (qSDV) with line mixing (qSDV+LM), together with spectroscopic line parameters from Devi et al. [1,2] for the 2v(3) band of CH4, was used to retrieve total columns of CH4 front atmospheric solar absorption spectra. The qSDV line shape (Tran et al., 2013) [3] with line mixing (Levy et al., 1992) [4] was implemented into the forward model of GFIT (the retrieval algorithm that is at the heart of the GGG software (Wunch et al., 2015) [5]) to calculate CH4 absorption coefficients. High-resolution laboratory spectra of CH4 were used to assess absorption coefficients calculated using a Voigt line shape and spectroscopic parameters from the atm line list (Toon, 2014) [6]. The same laboratory spectra were used to test absorption coefficients calculated using the qSDV+LM line shape with spectroscopic line parameters from Devi et al. [1,2] for the 2v(3) band of CH4 and a Voigt line shape for lines that don't belong to the 2v3 band. The spectral line list for lines that don't belong to the 2v(3) band is an amalgamation of multiple spectral line lists. We found that for the P, Q, and R branches of the 2v(3) band, the qSDV+LM simulated the laboratory spectra better than the Voigt line shape. The qSDV+LM was also used in the spectral fitting of high-resolution solar absorption spectra from four ground-based remote sensing sites and compared to spectra fitted with a Voigt line shape. The average root mean square (RMS) residual for 131,124 solar absorption spectra fitted with absorption coefficients calculated using the qSDV+LM for the 2v(3) band of CH4 and the new Spectral line list for lines for lines that don't belong to the 2v(3) band, was reduced in the P, Q and R branches by 5%, 13%, and 3%, respectively when compared with spectra fitted using a Voigt line shape and the atm line list. We found that the average total column of CH4 retrieved from these 131,124 spectra, With the qSDV+LM was 1.1 +/- 0.3% higher than the retrievals perfo
Service-oriented computing is revolutionizing the modern computing paradigms with its aim to boost software reuse and enable business agility. Under this paradigm, new services are fabricated by composing available se...
详细信息
Service-oriented computing is revolutionizing the modern computing paradigms with its aim to boost software reuse and enable business agility. Under this paradigm, new services are fabricated by composing available services. The problem arises as how to effectively and efficiently compose heterogeneous services facing the high complexity of service composition. Based on environment ontology, this paper introduces a requirement-driven service composition approach. We propose the algorithms to decompose the requirement, the rules to deduct the relation between services, and the algorithm for composing service. The empirical results and the comparison with other services' composition methodologies show that this approach is feasible and efficient.
Texture and color information as important characters of video objects have been widely used in the detection of moving objects. A detection algorithm based on texture may cause detection errors in regions of blank te...
详细信息
Texture and color information as important characters of video objects have been widely used in the detection of moving objects. A detection algorithm based on texture may cause detection errors in regions of blank texture and heterogeneous texture, and a detection algorithm based on color is easily influenced by illumination changes and shadows. In this paper, a new detection and fusion algorithm is proposed. At the detection stage based on texture, the background texture is classified according to the steering kernel. At the fusion stage, for the moving objects detected on the basis of texture and color respectively, a scheme based on a boundary selection strategy is proposed for combining the different detection objects. A relatively smooth boundary is selected as the true boundary, and the shadow detection is carried out to assist the boundary selection. Experimental results verify the advantages of the proposed algorithm as compared to the existing state-of-the-art algorithms. (C) 2012 Elsevier Ltd. All rights reserved.
暂无评论