摘要:
Document type comparison and classification using layout classification is accomplished by first segmenting a document page into blocks of text and white space. A grid of rows and columns, forming bins, is created on the page to intersect the blocks. Layout information is identified using a unique fixed length interval vector, to represent each row on the segmented document. By computing the Manhattan distance between interval vectors of all rows of two document pages and performing a warping function to determine the row to row correspondence, two documents may be compared by their layout. Furthermore, interval vectors may be grouped into N clusters with a cluster center, defined as the median of the interval vectors of the cluster, replacing each interval vector in its cluster. Using Hidden Markov Models, documents can be compared to document type models comprising rows represented by cluster centers and identified as belonging to one or more document types. In addition, documents stored in a database may be retrieved, deleted, or otherwise managed by type, using their corresponding vector sets without requiring expensive OCR of the document. Furthermore, based on the classification, it is a simple matter to locate which blocks of data contain certain information. Where only that information is desired, it is not necessary to perform OCR on the entire document. Rather OCR may be limited to those blocks where the particular information is expected based on the document type.
摘要:
A method for comparing a ranking scheme for an object, such as a document image, with a known standard, or ideal ranking scheme for a comparative assessment of the performance and reliability of the two ranking schemes, is taught. The same set of objects are processed and ranked by both ranking schemes. The ranking scheme being tested is examined for all subject objects ranked lower than its ranking by the ideal ranking scheme. The test ranking scheme is charged for each swapped object which it ranked higher than the subject object, which the ideal ranking scheme ranked lower.
摘要:
In a process of signature verification, an off-line test signature is compared to one or more on-line reference signatures. Dynamic data obtained from the reference signature or signatures may be used to improve the thinning of the test signature. Such data may also be used for segmenting the test signature and sequencing the resulting strokes. In some embodiments, the invention includes evaluating a dynamic error, based at least in part on such sequencing of the test signature. In some further embodiments, the invention includes producing a rendering of at least one on-line reference signature as a two-dimensional image, and evaluating a static error based, at least in part, on shape-matching between the rendered image and the test signature.
摘要:
A method and apparatus for scheduling the transmission of data packets over a multihop wireless backhaul network in which delay guarantees through the network may be advantageously ensured. Illustratively, a novel packet scheduling scheme is provided which is advantageously based on an existing scheduling policy for wireline networks and for which a delay guarantee (based on the delay characteristics of the existing scheduling policy) can be advantageously ensured. In one embodiment, an even-odd link activation framework is defined for a given multihop wireless backhaul network, and an associated scheduling policy based on an arbitrary existing scheduling policy for wireline networks is derived and adopted. In such a case, the derived scheduling policy, when applied to the given multihop wireless backhaul network, advantageously ensures a worst-case delay guarantee of approximately twice that of the existing scheduling policy for wireline networks.
摘要:
A system and method for processing and managing electronic copies of handwritten notes are provided. The system includes a processor for processing the handwritten notes to generate the electronic copies, with each electronic copy associated with a respective identifier corresponding to at least one set of the respective handwritten notes, in which the identifiers facilitate the management of the electronic copies. The system includes an electronic notepad and can also include devices operatively connected to the electronic notepad for operating with the electronic notepad to receive, manage, merge, and/or display the electronic copies from the electronic notepad.
摘要:
A method and apparatus for grooming traffic demands according to mileage based tariffs. An Integer Linear Program (ILP) that captures the traffic grooming problem is defined, and such a linear program can in principle be solved by conventional linear program application systems which are fully familiar to those of ordinary skill in the art. However, the time required to solve such an ILP is fairly large, even for the moderately sized networks we are interested in. That is, there are many possible routes to consider, and hence many integer variables in the ILP. Therefore, further in accordance with the principles of the present invention, the ILP is advantageously run on the Delaunay Triangulation of the network rather than on the completely connected network graph.
摘要:
The present invention provides a wide-sense, non-blocking WDM cross-connect that utilizes a relatively small number of wavelength interchangers. The cross-connect of the present invention comprises first, second and third fabrics and one or more wavelength interchangers that interconnect the first and second fabrics. Demands that require a change in wavelength are routed through the first fabric, through one or more of the wavelength interchangers, and through the second fabric. Demands that do not require a change in wavelength are only routed through the third fabric. A routing algorithm is utilized to ensure that any sequence of connection requests and withdrawals can be routed without disturbing any currently routed requests.
摘要:
The present invention provides a strictly non-blocking WDM cross-connect that utilizes a relatively small number of wavelength interchangers. In accordance with one embodiment, the WDM cross-connect utilizes n1k1 wavelength interchangers, where n1 is an integer equal to the number of wavelengths carried on an input optical fiber associated with the WDM cross-connect and k1 is the total number of input optical fibers to the WDM cross-connect. Each of the wavelength interchangers is connected to exactly one input port of the fabric and each input port of the fabric is connected to exactly one wavelength interchanger. The input optical fibers of the WDM cross-connect are connected to k1 optical switches of the cross-connect that separate out the n1 wavelengths onto n1 optical fibers, which are input to respective wavelength interchangers. Therefore, each wavelength interchanger receives exactly n1 optical fibers. The fabric has output ports that are connected to k2 output optical fibers, where k2 is an integer equal to the number of output optical fibers that are connected to the output ports of the fabric. In accordance with another embodiment of the present invention, the WDM cross-connect comprises two fabrics and one or more wavelength interchangers that interconnect the fabrics. The cross-connect in accordance with this embodiment can be rendered strictly non-blocking by utilizing a number of wavelength interchangers equal to (k1+k2)−1, where k1 and k2 correspond to the number of input and output optical fibers, respectively, of the WDM cross-connect. In accordance with both of these embodiments, the number of input and output optical fibers of the strictly non-blocking WDM cross-connect can be unequal.
摘要:
A system and method are disclosed for time-domain wavelength interleaved networking that reduce the need for complex time-slot scheduling and reduce the routing complexity. Substantially all communications in the time-domain wavelength interleaved network pass through a hub node. In addition, interior nodes in the time-domain wavelength interleaved network will forward substantially all communications received from the hub node that are destined for another node on all branches outward from the hub node. The central hub node can impose a timing reference. Thus, the transmission and reception of a message can be synchronized such that a message sent in a time-slot k by a node Ni will be received by a node Nj in the time-slot k. Further, the hub node can recover from a link failure by shifting transmission times of all nodes that are separated from the hub node by the failed link.
摘要:
A method for exchanging routing information between I-BGP routers within an autonomous system (AS) advantageously enables a solution to both persistent route oscillation problems and transient route oscillation problems which may occur when using I-BGP in a given AS. Conventional I-BGP protocol techniques are extended by enabling I-BGP speakers (e.g., routers) to communicate a set of possible best paths to a given destination, rather than communicating only a single best path, to each of their I-BGP peers within the given AS. Specifically, a plurality of possible best paths to a destination are communicated (where there are in fact more than one) from an I-BGP speaker in a given AS to its I-BGP peers (within the given AS), for each neighboring AS that provides any such paths (i.e., routes to the destination).