In this paper we propose a technique to incorporate contextual information into object classification. In the real world there are cases where the identity of an object is ambiguous due to the noise in the measurement...
In this paper we propose a technique to incorporate contextual information into object classification. In the real world there are cases where the identity of an object is ambiguous due to the noise in the measurements based on which the classification should be made. It is helpful to reduce the ambiguity by utilizing extra information referred to as context, which in our case is the identities of the accompanying objects. This technique is applied to white blood cell classification. Comparisons are made against "no context" approach, which demonstrates the superior classification performance achieved by using context. In our particular application, it significantly reduces false alarm rate and thus greatly reduces the cost due to expensive clinical tests.
Recent advances in technology make it possible to integrate multiple processors into a single chip to build high performance parallel programmable digital signal processors (PPDSPs). These processors are expected to r...
详细信息
Recent advances in technology make it possible to integrate multiple processors into a single chip to build high performance parallel programmable digital signal processors (PPDSPs). These processors are expected to replace many dedicated digital signal processors to implement important image/signal processing algorithms such as discrete cosine transform (DCT). The paper addresses the issue of how to compare fast 2D-DCT algorithms when they are implemented on a PPDSP. Previously, the efficiency of these algorithms is compared based on the number of operations. This comparison is reasonable when these algorithms are implemented on a dedicated DSP. However, this comparison may not be suitable for general-purpose PPDSPs. The paper proposes to use three parameters, the number of data accesses, the number of communications, and the distance of communications, as new criterion for performance comparison of DCT algorithms. An algorithm-level technique is developed to estimate these parameters for DCT algorithms. The comparison results based on the parameters show that the algorithm proposed by Cho and Lee (1991) might be the best choice for a PPDSP unless it requires large overhead for communication between remote processors. In this case, the conventional row-column method with a fast 1D-DCT algorithm might be the most efficient.
We review our recent research in the field of photochromic polymeric structures for optical data storage and nonlinear optics. We unify some of the sub -Themes of azopolymer structures in the light of photo -induced m...
详细信息
A jamming game is considered. A transmitter encodes the data, in a product code (PC), and transmits it in a slow frequency-hopped spread-spectrum (FHSS) system. A hostile jammer attempts to block correct packet recept...
详细信息
A jamming game is considered. A transmitter encodes the data, in a product code (PC), and transmits it in a slow frequency-hopped spread-spectrum (FHSS) system. A hostile jammer attempts to block correct packet reception by adjusting the fraction of the total band in which it distributes the jam power, /spl rho/. Two cases are considered. In the first case, the transmitter declares his strategy and the jammer uses this information to select his best jamming fraction. In the second case, the jammer declares his strategy first and the transmitter uses this information to select his best rates. Results are obtained using asymptotic analysis of a product code using singly extended Reed-Solomon (RS) row and column codes whose size grows large. This asymptotic analysis leads to good choices of the row and column rates for the PC in both cases.
A parallel method developed for reconstructing images from holograms obtained with optical scanning holography is presented. A standard portable message-passing library, MPI or message-passing interface, is used to pr...
详细信息
A parallel method developed for reconstructing images from holograms obtained with optical scanning holography is presented. A standard portable message-passing library, MPI or message-passing interface, is used to provide the environment of parallel programming. The parallel method is implemented using a network of workstations. The process of reconstructing images with twin-image removal from holograms is performed in parallel with respect to different reconstruction planes of interest. This process contributes to possible faster three-dimensional object identification to be investigated in future research.
It is generally agreed that the primary scientific goal of the first near Sun fly-by mission is to identify the source of momentum responsible for driving the solar wind and to understand the physics of solar coronal ...
It is generally agreed that the primary scientific goal of the first near Sun fly-by mission is to identify the source of momentum responsible for driving the solar wind and to understand the physics of solar coronal heating. In this paper, we present an innovative concept for obtaining a unique set of data with which to better understand the physics of solar wind acceleration and coronal heating. The instrument, called the Solar Event Correlation Analyzer (SECA), operates on the principle that the topology of the open magnetic field lines can be used to causally correlate the evolution of small scale structures (events) in the Sun’s transition region with in situ solar probe measurements. These small-scale events are generally viewed as the mechanisms providing energy and momentum for the solar corona and wind, an assumption that will be conclusively tested with the SECA instrument. This paper describes the SECA concept and its scientific justification.
In an era of fiscal austerity, downsizing and unforgiving pressure upon human and economic capital, it is an Augean task to identify resources for fresh and creative work. The realities of the day and the practical de...
详细信息
In an era of fiscal austerity, downsizing and unforgiving pressure upon human and economic capital, it is an Augean task to identify resources for fresh and creative work. The realities of the day and the practical demands of more immediate fleet needs can often dictate higher priorities. Yet, the Navy must avoid eating its seed corn. Exercising both technical insight and management foresight, the fleet, the R&D community, the Office of the Chief of Naval Operations (OpNav) and the product engineering expertise of the Naval Surface Warfare Center (NSWC) are joined and underway with integrated efforts to marry new, fully demonstrated technologies and operational urgencies. Defense funding today cannot sponsor all work that can be mission-justified over the long term because budgets are insufficient to support product maturation within the classical development cycle. However, by rigorous technical filtering and astute engineering of both marketplace capabilities and currently available components, it is possible in a few select cases to compress and, in effect, integrate advanced development (6.3), engineering development (6.4), weapon procurement (WPN), ship construction (SCN), operation and maintenance (O&M,N) budgetary categories when fleet criticalities and technology opportunities can happily meet. In short, 6.3 funds can be applied directly to ''ripe gateways'' so modern technology is inserted into existing troubled or aging systems, sidestepping the lengthy, traditional development cycle and accelerating practical payoffs to recurrent fleet problems. To produce such constructive results has required a remarkable convergence of sponsor prescience and engineering workforce excellence. The paper describes, extensively, the philosophy of approach, transition strategy, polling of fleet needs, technology assessment, and management team requirements. The process for culling and selecting specific candidate tasks for SHARP sponsorship (matching operational need with t
Using the finite-difference time-domain method, three-dimensional scattering patterns are computed for cells containing multiple organelles. The scattering cross section and average cosine of the scattering angle are ...
详细信息
Using the finite-difference time-domain method, three-dimensional scattering patterns are computed for cells containing multiple organelles. The scattering cross section and average cosine of the scattering angle are computed for cells as a function of volume fraction of melanin granules and mitochondria, Results show that small organelles play a significant role in light scattering from cells, and the volume fraction of organelles affects both the total amount of scattered light and the angular distribution of scattered light.
The relationship between optical properties and image contrast in confocal imaging is investigated. A Monte Carlo simulation has been developed to analyze the effects of changes in scattering, index of refraction, and...
详细信息
The relationship between optical properties and image contrast in confocal imaging is investigated. A Monte Carlo simulation has been developed to analyze the effects of changes in scattering, index of refraction, and absorption in a three-layer medium. Contrast was calculated from the computed signal-to-background ratios for changes in tissue optical properties. Results show that the largest source of contrast is changes in refractive index. (C) 1996 Optical Society of America
暂无评论