Pattern recognition neural network with saccade-like operation
First Claim
1. A multi-layered neural network for pattern recognition, comprising:
- an input layer for mapping into a two-dimensional scan window, said scan window defining an input space that can be scanned over objects, and which input contains at least two adjacent ones of said objects;
an output layer comprised of at least a first output node with an associated activation energy that is activated to represent the presence of a desired object substantially centered in said scan window, the activation energy thereof having a variable level, which level represents the proximity of the object to the substantial center of said scan window, and a second output node with an associated activation energy that is activated to represent a desired distance from a frame of reference in said scan window to another and adjacent one of said objects when the activation energy thereof is raised above a predetermined threshold; and
a hidden layer having local receptor fields and interconnected with said input layer and said output layer for mapping said input layer to said output layer, said hidden layer providing a representation of the position of the desired one of said objects relative to the substantial center of said scan window such that said first output node is activated in response thereto, said hidden layer providing a representation of the desired distance when the adjacent one of said objects is separated from said frame of reference in said scan window by substantially said desired distance such that said second output node is activated in response thereto.
2 Assignments
0 Petitions
Accused Products
Abstract
A multi-layered pattern recognition neural network (30) is disclosed that comprises an input layer (50) that is operable to be mapped onto an input space that includes a scan window (32). Two hidden layers (54) and (58) map the input space to an output layer (34). The hidden layers utilize a local receptor field architecture and store representations of objects within the scan window (32) for mapping into one of a plurality of output nodes. Further, the output layer (34) is also operable to store representations of desired distances between the center of the scan window (32) and the next adjacent object thereto and also the distance between the center of the scan window (32) and the center of the current object. A scanning system can then utilize the information regarding the distance to the next adjacent object, which is stored in an output vector (40) to incrementally jump to the center of the next adjacent character rather than scan the entire distance therebetween. This is referred to as a saccade operation. Once the scan window ( 32) is disposed over the next object, a corrective saccade can be performed by utilizing the information output by the neural network (30) relating to the distance between the center of the scan window (32) and the current character. This information is output as an output vector (38) from the neural network (30).
195 Citations
39 Claims
-
1. A multi-layered neural network for pattern recognition, comprising:
-
an input layer for mapping into a two-dimensional scan window, said scan window defining an input space that can be scanned over objects, and which input contains at least two adjacent ones of said objects; an output layer comprised of at least a first output node with an associated activation energy that is activated to represent the presence of a desired object substantially centered in said scan window, the activation energy thereof having a variable level, which level represents the proximity of the object to the substantial center of said scan window, and a second output node with an associated activation energy that is activated to represent a desired distance from a frame of reference in said scan window to another and adjacent one of said objects when the activation energy thereof is raised above a predetermined threshold; and a hidden layer having local receptor fields and interconnected with said input layer and said output layer for mapping said input layer to said output layer, said hidden layer providing a representation of the position of the desired one of said objects relative to the substantial center of said scan window such that said first output node is activated in response thereto, said hidden layer providing a representation of the desired distance when the adjacent one of said objects is separated from said frame of reference in said scan window by substantially said desired distance such that said second output node is activated in response thereto. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12)
-
-
13. A multi-layered neural network system for pattern recognition, comprising:
-
an input field containing a plurality of adjacently disposed objects; a scanning mechanism for generating a two-dimensional scan window and scanning said scan window over said objects such that said scan window contains at least two of said adjacent objects; an input layer for mapping into said two-dimensional scan window; an output layer comprised of at least a first output node with an associated activation energy that is activated to represent the presence of a desired one of said objects substantially centered in said scan window, the activation energy thereof, and a second output node with an associated activation energy that is activated to represent a desired distance from a frame of reference in said scan window to another and adjacent one of said objects when the activation energy is raised above a predetermined threshold; a hidden layer having local receptor fields and interconnected with said input layer and said output layer for mapping said input layer to said output layer, said hidden layer providing a representation of this position of the desired object relative to the substantial center of said scan window such that said first output node is activated in response thereto, said hidden layer providing a representation of the desired distance when said adjacent object is separated from said frame of reference in said scan window by substantially said desired distance such that said second output node is activated in response thereto; and a processor for receiving the output of said first output node and recognizing when the first output node is activated indicating the presence of said desired object centered in said scan window and providing an output in response thereto, and said processor operable to receive said second output node and controlling said scanning mechanism to move by a distance substantially corresponding to said associated desired distance when said second node is activated. - View Dependent Claims (14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24)
-
-
25. A method for recognizing a pattern, comprising:
-
providing an input layer in a neural network; mapping the input layer into a two dimensional scan window, the scan window defining an input space that can be scanned over objects, and which scan window can contain at least two adjacent ones of the objects; providing an output layer in the neural network having at least a first output node with an associated activation energy that is operable to be activated to represent the presence of a desired one of the objects relative to the substantial center of the scan window and a second output node with an associated activation energy that is activated to represent a desired distance between the frame of reference within the scan window and the one of the objects not substantially centered in the scan window; the first and second output nodes activated when the associated activation energies rise above respective predetermined thresholds; providing a hidden layer in the neural network and interconnecting the hidden layer with the input layer and the output layer; mapping the input layer to the output layer with the hidden layer, the hidden layer providing a representation of the position of the desired one of the objects relative to the substantial center of the scan window, and providing a representation of the desired distance when the desired one of the objects not substantially centered in the scan window is separated from the frame of reference by substantially the desired distance; activating the first output node in response to the desired one of the objects being disposed within the scan window and substantially corresponding to the representation in the hidden layer; and activating the second output node when the distance from the frame of reference to the one of the objects not substantially centered in the scan window is substantially equal to the desired distance. - View Dependent Claims (26, 27, 28, 29, 30, 31, 32, 33)
-
-
34. A method for recognizing a pattern, comprising:
-
providing an input layer in a neural network; mapping the input layer into the scan window; providing an output layer in the neural network having at least a first output node with an associated activation energy that is activated to represent the presence of a desired object in the scan window and the position thereof relative to the substantial center of the scan window and a second output node with an associated activation energy that is activated to represent a desired distance between the frame of reference within the scan window and the one of the objects not substantially centered in the scan window; providing a hidden layer in the neural network and interconnecting the hidden layer with the input layer and the output layer; mapping the input layer to the output layer with the hidden layer, the hidden layer providing a representation of the position of the desired object relative to the substantial center of the scan window, and providing a representation of the desired distance when the object not substantially centered in the scan window is separated from the frame of reference by substantially the desired distance; activating the first output node by raising its associated activation energy to a level representing the position of the desired object relative to the substantial center of the scan window in response to the desired object being disposed in of the scan window and substantially corresponding to the representation in the hidden layer; activating the second output node by raising its associated activation energy above a predetermined threshold when the distance from the frame of reference to the one of the objects not substantially centered in the scan window is substantially equal to the desired distance; and moving the scan window by a distance substantially corresponding to the desired distance associated with the second output node after activation of both the first and second output nodes. - View Dependent Claims (35, 36, 37, 38, 39)
-
Specification