Showing papers by "Google published in 1995"

Comparison of learning algorithms for handwritten digit recognition

Abstract: COMPARISON OF LEARNINGALGORITHMS FOR HANDWRITTEN DIGITRECOGNITIONY. LeCun, L. Jackel, L. Bottou, A. Brunot, C. Cortes,J. Denker, H. Drucker, I. Guyon, U. M uller,E. Sackinger, P. Simard, and V. VapnikBell Lab oratories, Holmdel, NJ 07733, USAEmail: yann@research.att.comAbstractThis pap er compares the p erformance of several classi er algorithmson a standard database of handwritten digits. We consider not only rawaccuracy, but also rejection, training time, recognition time, and memoryrequirements.1

Learning algorithms for classification: A comparison on handwritten digit recognition

Abstract: This paper compares the performance of several classi er algorithms on a standard database of handwritten digits. We consider not only raw accuracy, but also training time, recognition time, and memory requirements. When available, we report measurements of the fraction of patterns that must be rejected so that the remaining patterns have misclassi cation rates less than a given threshold.

Combined dictionary based and likely character string method of handwriting recognition and display

Abstract: In a handwriting recognition process, a list of candidate recognized words is identified (202) as a function of both comparison of dictionary entries to various combinations of recognized character combinations, and through a most likely character string and most likely string of digits analysis as developed without reference to the dictionary. The process selects (301) a word from the list and presents (302) this word to the user. The user then has the option of displaying (303) this list. When displaying the list, candidate words developed with reference to the dictionary are displayed in segregated manner from the most likely character string words and the most likely string of digits. The user can charge the selected word by choosing from the list, or edit the selected word. When the user selects the most likely character string as the correct representation of the handwritten input to be recognized, the process automatically updates (310) the dictionary to include the most likely character string The same process can occur when the user selects the most likely string of digits.

Method and apparatus for performing handoff in a packet data communication system

Abstract: A packet data communication system (100) that includes a packet data routing network (101, 103) and a plurality of base sites (105-107) employs a method (400, 500) and apparatus for providing packet data to a communication unit (109). The packet data routing network (101, 103) provides a packet (201-208) of data to at least two of the base sites (105-107), one of the base sites (e.g., 105) being the site currently serving the communication unit (109) and the other base site or sites (e.g., 106, 107) being possible handoff targets. The serving base site (105) transmits at least a first portion (e.g., 201-203) of the packet (201-208) to the communication unit (109) and the communication unit (109) receives the transmitted packet or portion thereof. However, if during transmission of the packet (201-208) the communication unit (109) requires a handoff to one (e.g., 106) of the target base sites (106, 107), the communication unit (109) transmits an acknowledgment message (124) to the target base site (106) indicating receipt of the first portion (201-203) of the packet. Upon receipt of the acknowledgment message (124), the target base site (106) transmits at least a second portion (e.g., 204-208) of the packet to the communication unit (109).

Method and apparatus for providing navigation guidance

Abstract: A method and apparatus (10) for providing navigation guidance via a land vehicle by providing a navigation route planner (24) having a road network database (12) for planning a navigation route along the road network, determining a current estimated position of the land vehicle, communicating maneuver instructions to a driver of the land vehicle at a predetermined maneuver distance base value before the land vehicle reaches a required maneuver along the planned navigation route, and detecting and quantifying current environmental travel conditions for assisting a navigation device in guiding the land vehicle along the planned navigation route.

Optoelectronic interconnect device and method of making

Abstract: A waveguide (12) having a core region (16) and a cladding region (20) is formed. A portion of the cladding region (20) forms a first surface (21) and portions of both the core region (16) and the cladding region (20) forms an end surface (15). An insulative flexible substrate (14) having an electrically conductive tracing (32) with a first portion and a second portion, wherein the first portion of the insulative flexible substrate (14) is mounted on the end surface of the waveguide (12).

Method and system for extracting features from handwritten text

Abstract: A handwriting recognition system achieves a higher recognition rate by using a feature extraction method which computes features based on multiple data frames A plurality of data frames is generated from handwritten text received by the system (42) Each data frame includes samples taken from the handwritten text (44) Individual-frame features are extracted from individual data frames (46), and in turn, multi-frame features are extracted from individual-frame features (48) which correspond to different data frames

Method and system for recognizing a boundary between characters in handwritten text

Abstract: A method and system for identifying boundaries of characters in handwritten text by classifying segment strokes provides improved performance in a handwriting recognition system. A segment stroke is a portion of handwritten text which includes a boundary between two characters. The segment stroke is recognized by the same method used to recognize characters. Recognition of a segment stroke is accomplished by training a learning machine to act as a classifier (208) which implements a discriminant function based on a polynomial expansion.

Alkaline gel electrolyte for electrochemical cells

Abstract: An alkaline gel electrolyte for an electrochemical cell is based on a polymeric binder of polyvinyl alcohol or polyvinyl acetate. Potassium hydroxide and/or sodium hydroxide is blended into the polymeric binder or matrix, and the resulting solution is used to cast a film that is useful as a gel electrolyte. The hydroxide in the solution is between about 1% to 25% by weight, and the polymer concentration is between about 0.5% to 10% by weight. The resulting gel electrolyte film contains between 10 to 90% water by weight. A nickel or iron porphine modifier may be added in an amount between about 1 to 1000 parts per million. The alkaline gel electrolyte is useful in combination with asymmetric inorganic electrodes to make electrochemical cells, and can also function as a separator. Improved power density and higher cycle life is achieved with the gel electrolyte.

Method and apparatus for synchronization of a communications unit

Abstract: The disclosed communication unit includes synchronization circuitry having a sync sequence match detector and a decode detector. Upon matching a received sync sequence or word with a known sync sequence, the sync word (or, e.g., the block containing the received sync sequence) is decoded to determine whether the block is correctable (or more generally, whether the signal quality is above a threshold). When a match is found and the sequence/block is correctable, an appropriate sync indication is outputted. Additional information, such as color code matching of consecutive sync words, may also be used before determining synchronization has been achieved.

Implanted LED array and method of fabrication

Abstract: A method of fabricating an LED array including epitaxially and sequentially growing a conductive layer (12, 12') on a substrate (10, 10'), a first carrier confinement layer (13, 13'), an active layer (14, 14'), a second carrier confinement layer (15, 15') and a conductive cap (16, 16') selectively formed to provide exposed surface areas (17, 17') defining row and column areas with a matrix of diodes positioned in rows and columns therebetween. Implanting a first impurity (20, 20') in the row areas to form vertical conductors extending through the second confinement, active and first confinement layers to provide surface contacts to each diode. Implanting a second impurity (25, 25') in the row and column areas through the second confinement and active layers to form an isolating resistive volume around each diode. Implanting a third impurity (30, 30') in the row areas through the second confinement, active, and first confinement layers and into the substrate to form an isolating resistive volume between each row of diodes.