Index

A B C D E F G H I K L M N O P R S T U V W

A

AboutFrame - Class in <Unnamed>: Displays a simple 'about' dialog referencing the author.
AboutFrame(Frame) - Constructor for class AboutFrame: Creates a new about frame.
accept(File) - Method in class ImageFileFilter: Returns true or false to indicate acceptance of the file.
accept(File) - Method in class NetFileFilter: Returns true or false to indicate acceptance of the file.
accept(File) - Method in class SymbolFileFilter: Returns true or false to indicate acceptance of the file.
actionPerformed(ActionEvent) - Method in class LoadFrame: Is called whenever a component fires an actionEvent.
actionPerformed(ActionEvent) - Method in class NetPanel: Is called whenever an appropriate event is fired by a component on the panel, probably a button.
actionPerformed(ActionEvent) - Method in class NewNetFrame: Is called whenever a component fires an actionEvent.
actionPerformed(ActionEvent) - Method in class OCRSwing: Fired when a component performs an action whilst this class is listening to it.
actionPerformed(ActionEvent) - Method in class ReceptorPanel: Is called whenever an appropriate event is fired by a component on the panel, probably a button.
actionPerformed(ActionEvent) - Method in class ResultsPanel: Fired when a component performs an action whilst this class is listening to it.
actionPerformed(ActionEvent) - Method in class RunPanel: Fired when a component performs an action whilst this class is listening to it.
actionPerformed(ActionEvent) - Method in class SymbolPanel: Fires when a component to which this object is attached fires an actionEvent.
actionPerformed(ActionEvent) - Method in class TrainPanel: Fired when a component performs an action whilst this class is listening to it.
add(Receptor) - Method in class NearOCR.ReceptorManager: Adds a single receptor to the set.
add(Symbol) - Method in class NearOCR.SymbolTable: Adds a symbol to the end of the symbol table.
addEdge(Edge) - Method in class NearNeural.ReceptiveNeuron: Adds a single edge to the neuron.
addEdges(Neuron[]) - Method in class NearNeural.ReceptiveNeuron: Constructs randomly weighted edges to all neurons in the array provided.
addLetter(Letter) - Method in class NearOCR.Document: Adds a letter to the list of letters.
addMessageListener(MessageListener) - Static method in class NearOCR.Logger: Adds a message listener, which will from that moment on receive all logged events.
addNeuron(Neuron) - Method in class NearNeural.NeuronLayer: Adds a neuron to the layer.
addSymbol(Symbol) - Method in class NearOCR.OCRManager: Adds a symbol to the symbol table.
adjustmentValueChanged(AdjustmentEvent) - Method in class SlicePanel: Handles a change in the scrolling pane, redrawing the image each time the user drops the scrollbar.
adjustSymbolWeights(NeuralNet[], Symbol, ProgressListener, ReceptorManager) - Static method in class NearOCR.TrainingManager: Adjusts all symbol weights in a single symbol.
adjustWeightByDeviation(double) - Method in class NearOCR.WeightedNode: Adjusts the weight of this node such that its weight multiplied by the given output is exactly equal to the target.
analyse(double, double, ProgressListener) - Method in class NearOCR.OCRManager: Analyses all letters by running them through all nets.
analyseSymbolWeights(ProgressListener) - Method in class NearOCR.OCRManager: Analyses and updates all symbol weights to compensate for individual net idiosynchrasies.
appendNet(NeuralNet) - Method in class NearOCR.NetStack: Appends a net to the end of the stack.
appendNet(NeuralNet) - Method in class NearOCR.OCRManager: Appends a net to the stack.

B

backPropagateHidden(NeuronLayer, ReceptiveNeuron[]) - Method in interface NearNeural.BackPropagationMethod: Backpropagates a Hidden layer in the net.
backPropagateHidden(NeuronLayer, ReceptiveNeuron[]) - Method in class NearNeural.ThreadedUnweightedBackPropagationMethod: Back propagates a hidden layer relative to the layer immediately following it.
backPropagateHidden(NeuronLayer, ReceptiveNeuron[]) - Method in class NearNeural.UnweightedBackPropagationMethod: Back propagates a hidden layer relative to the layer immediately following it.
backPropagateOutputs(double[], ReceptiveNeuron[]) - Method in interface NearNeural.BackPropagationMethod: Backpropagates the last layer directly based on the desirable set given.
backPropagateOutputs(double[], ReceptiveNeuron[]) - Method in class NearNeural.ThreadedUnweightedBackPropagationMethod: This function back-propogates output nodes.
backPropagateOutputs(double[], ReceptiveNeuron[]) - Method in class NearNeural.UnweightedBackPropagationMethod: This function back-propogates output nodes.
BackPropagationMethod - Interface in NearNeural: Allows formally-specified backpropagation methods to be defined in order to modify the training of the net.

C

calculateValue() - Method in class NearNeural.ReceptiveNeuron: pre-calculates the value of this neuron - this method requires that all previous layers have been run first in a feed-forward manner.
CharacterImage - Class in NearOCR: A bitmap representation of a character.
CharacterImage(BufferedImage) - Constructor for class NearOCR.CharacterImage: Creates a new CharacterImage from a BufferedImage.
clearLetters() - Method in class NearOCR.Document: Clears the document of letters.
ConsolePanel - Class in <Unnamed>: A panel that dsplays the logger results for NearOCR.
ConsolePanel() - Constructor for class ConsolePanel: Creates a new console panel, attaching itself to the logger and placing the log panel.
count() - Method in class NearOCR.ReceptorManager: Returns the number of receptors in the set.
countEdges() - Method in class NearNeural.ReceptiveNeuron: Returns the number of edges currently in this neuron.
countInputs() - Method in class NearOCR.OCRManager: Returns the number of inputs as shared across the net stack.
countNets() - Method in class NearOCR.NetStack: Returns the number of nets that are currently loaded.
countNeurons() - Method in class NearNeural.NeuronLayer: Returns the number of neurons currently in this layer.
countOutput() - Method in class NearOCR.NetStack: Returns the number of outputs this OCR system will generate.
countOutput() - Method in class NearOCR.OCRManager: Returns the number of outputs in the final layer of the entire net stack.
createNewLayer() - Method in class NewNetFrame: Creates a new layer and adds it to the JList, but not to the OCRManager or the provisional net.

D

delta - Variable in class NearNeural.ReceptiveNeuron: Stores the last value correction as per the backpropagation algorithm
differentialValue(Edge[]) - Method in class NearNeural.HyperbolicTangentThresholdModel: The value as run through the differential of the current function.
differentialValue(Edge[]) - Method in class NearNeural.LinearThresholdModel: Returns the differential of the thresholding function.
differentialValue() - Method in class NearNeural.ReceptiveNeuron: Returns the differential of the function that generates value at this point.
differentialValue(Edge[]) - Method in class NearNeural.SigmoidThresholdModel: Defined in ThresholdingAlgorithm, returns differential of current value.
differentialValue(Edge[]) - Method in class NearNeural.SinusoidalThresholdModel: Returns the differential of the thresholding function.
differentialValue(Edge[]) - Method in class NearNeural.ThreePointRuleDifferentiation: Attempts to find the differential of any function by way of numerical methods.
differentialValue(Edge[]) - Method in interface NearNeural.ThresholdingAlgorithm: Returns the first differential of the value.
Document - Class in NearOCR: Retains information specific to one document.
Document(BufferedImage) - Constructor for class NearOCR.Document: A simple constructor to instantiate the image on which the document is to be based.
DoubleObliqueReceptorPattern - Class in NearOCR: Provides a double-hatching layout designed to emulate pixel layouts.
DoubleObliqueReceptorPattern() - Constructor for class NearOCR.DoubleObliqueReceptorPattern
drawDocument() - Method in class SlicePanel: Draws the document with highlighted slices.
drawGlobalInfo() - Method in class NetPanel: Rcalculates and draws information about the global net situation.
drawNetInfo() - Method in class NetPanel: Recalculates then draws individual net information into the UI.
drawReceptors() - Method in class ReceptorPanel: Draws the document with highlighted slices.

E

Edge - Class in NearNeural: This represents an edge in a NearNeural net.
Edge(Neuron, double) - Constructor for class NearNeural.Edge: Creates a new edge with neuron and weight.
edges - Variable in class NearNeural.ReceptiveNeuron: Retains all Edges for this node.
errorOccurred(String) - Method in interface NearOCR.ProgressListener: called whenever an error, fatal or non-fatal, occurs during the operation.
errorOccurred(String) - Method in class RunPanel: Fired when an error occurs during the analysis process.
errorOccurred(String) - Method in class SymbolPanel: Fired when an error occurs during the analysis process.
errorOccurred(String) - Method in class TrainPanel: Fired when an error occurs during the analysis process.
eventOccurred(String) - Method in interface NearOCR.ProgressListener: Called whenever a noteworthy event during the operation completes.
eventOccurred(String) - Method in class RunPanel: Fired when any event occurs in the analysis cycle.
eventOccurred(String) - Method in class SymbolPanel: Fired when any event occurs in the analysis cycle.
eventOccurred(String) - Method in class TrainPanel: Fired when any event occurs in the analysis cycle.

F

FixedLengthRandomReceptorPattern - Class in NearOCR: Although slightly more capable than the random sampling receptor pattern this is still severely imperfect.
FixedLengthRandomReceptorPattern() - Constructor for class NearOCR.FixedLengthRandomReceptorPattern

G

generateReceptors(int) - Method in class NearOCR.DoubleObliqueReceptorPattern: An implementation of ReceptorPattern.generateReceptors(int).
generateReceptors(int) - Method in class NearOCR.FixedLengthRandomReceptorPattern: An implementation of ReceptorPattern.generateReceptors(int).
generateReceptors(int) - Method in class NearOCR.RandomReceptorPattern: An implementation of ReceptorPattern.generateReceptors(int).
generateReceptors(int) - Method in interface NearOCR.ReceptorPattern: Returns an array of Receptors number in length.
generateReceptors(int) - Method in class NearOCR.RegularObliqueReceptorPattern: An implementation of ReceptorPattern.generateReceptors(int).
getBackPropagationMethod() - Method in class NearNeural.NeuralNet: Returns the BackPropagationMethod used to train the net.
getDelta() - Method in class NearNeural.ReceptiveNeuron: Returns the last error value from backpropagation.
getDescription() - Method in class ImageFileFilter: Returns the description that is displayed in the drop down type selector in the FileChooser prompt.
getDescription() - Method in class NetFileFilter: Returns the description that is displayed in the drop down type selector in the FileChooser prompt.
getDescription() - Method in class SymbolFileFilter: Returns the description that is displayed in the drop down type selector in the FileChooser prompt.
getDesiredValues(int, int) - Method in class NearOCR.Symbol: Returns a range of desired values.
getDesiredValues() - Method in class NearOCR.Symbol: This returns a double array of all 'desirable' edges which should be firing in order to trigger the symbol.
getDeviation(double) - Method in class NearOCR.WeightedNode: Returns the difference between the weight-adjusted output neuron value and the internally stored target.
getDocument() - Method in class NearOCR.OCRManager: Returns the document that this OCRManager is using.
getEdge(int) - Method in class NearNeural.ReceptiveNeuron: Returns an edge at the index given.
getEdge(int, int) - Method in class NearNeural.ReceptiveNeuronLayer: Returns an edge from one of the neurons in this layer.
getHeight() - Method in class NearOCR.Letter: Returns the current height of the letter.
getImage() - Method in class NearOCR.CharacterImage: Returns the original, unscaled image which represents this symbol.
getImage() - Method in class NearOCR.Document: Returns the image that this document is based on.
getIndex() - Method in class NearOCR.WeightedNode: Returns the current index of this node.
getInputCount() - Method in class NearOCR.NetStack: Returns the number of input neurons that this OCR manager is currently set to using.
getInputEdges() - Method in class NearNeural.ReceptiveNeuron: Returns a list of all edges handled by this Neuron
getK() - Method in class NearNeural.WeightedThresholdModel: Returns the coefficient of the function.
getLayer(int) - Method in class NearNeural.NeuralNet: Returns a layer such that one may edit its nodes and structure manually.
getLayerStructure() - Method in class NearNeural.NeuralNet: Returns an array of integers representing the structure of the net.
getLearningRate() - Method in class NearNeural.UnweightedBackPropagationMethod: Returns the current learning rate.
getLength() - Method in class NearOCR.Symbol: Returns the number of weighted nodes held by this symbol.
getLength() - Method in class NearOCR.SymbolTable: Returns the length of the table.
getLetters() - Method in class NearOCR.Document: Returns an array of all Letters contained within this document.
getListCellRendererComponent(JList, Object, int, boolean, boolean) - Method in class NetCell: Provides a renderable component for the JList.
getListCellRendererComponent(JList, Object, int, boolean, boolean) - Method in class NeuronCell: Provides a renderable component for the JList.
getListCellRendererComponent(JList, Object, int, boolean, boolean) - Method in class SymbolCell: Provides a renderable component for the JList.
getMeanDeviation(double[]) - Method in class NearOCR.Symbol: Calculates a confidence value for this nsymbol based on net outputs.
getNet(int) - Method in class NearOCR.NetStack: Returns a net by index from this OCR manager.
getNet(int) - Method in class NearOCR.OCRManager: Returns a net from the given index number.
getNets() - Method in class NearOCR.NetStack: Returns an array of all nets held in this OCR manager, ordered as they are to have their outputs serialised.
getNets() - Method in class NearOCR.OCRManager: Returns all nets from the net stack, is merely a wrapper for the Net stack object.
getNeuron(int, int) - Method in class NearNeural.NeuralNet: Returns a neuron from anywhere in the current net.
getNeuron(int) - Method in class NearNeural.NeuronLayer: Returns a neuron with the index provided
getNeuron(int) - Method in class NearNeural.ReceptiveNeuronLayer
getNeurons() - Method in class NearNeural.NeuronLayer: Returns all Neurons in this layer.
getNeurons() - Method in class NearNeural.ReceptiveNeuronLayer: Returns all neurons as ReceptiveNeurons.
getNumberOfLayers() - Method in class NearNeural.NeuralNet: Returns the number of layers in the net, without any extra information.
getOldWeight() - Method in class NearNeural.Edge: Returns the old weight, as per backpropagation.
getOutput() - Method in class NearNeural.NeuralNet: Returns the output of the net as an array of doubles as long as the amount of output Neurons.
getReceptorManager() - Method in class NearOCR.OCRManager: Returns the current ReceptorManager that will be used for analysis.
getReceptors() - Method in class NearOCR.ReceptorManager: Returns an array with references to all receptors in the set, ordered by the order in which they were inserted.
getSource() - Method in class NearNeural.Edge: Retrns the target of this edge, ie the source (the neuron from which data flows unto the one which owns this Edge), but not the source of data.
getSupportedFileVersions() - Method in class NearNeural.NeuralNet: Returns an array of strings that represent (and correlate with) version ids from which this net can load files.
getSymbol(int) - Method in class NearOCR.SymbolTable: Returns a symbol by index.
getSymbols() - Method in class NearOCR.SymbolTable: Returns an array with all symbols from this table in.
getSymbolTable() - Method in class NearOCR.OCRManager: Returns the entire current sumbol table.
getTarget() - Method in class NearOCR.WeightedNode: Returns the current target value, between 0 and 1, of this weighted node.
getValue() - Method in class NearOCR.CharacterImage: Returns the value of this Symbol
getWeight() - Method in class NearNeural.Edge: Returns the weight of the Edge.
getWeight(int) - Method in class NearOCR.Symbol: Returns one weighted node, addressed by the index in the internal array - NOT the index stored within the weighted node!
getWeight() - Method in class NearOCR.WeightedNode: Returns the weight this node currently holds.
getWeights() - Method in class NearOCR.Symbol: Returns a list of all weighted nodes stored within this Symbol.
getWeightsFromOffset(int, int) - Method in class NearOCR.Symbol: Returns a range of weighted nodes from a given position.
getWidth() - Method in class NearOCR.Letter: Returns the current width of this letter.
getX() - Method in class NearOCR.Letter: Returns the current x-coordinate of the letter as placed on the document from which it was sliced.
getX1() - Method in class NearOCR.Receptor
getX2() - Method in class NearOCR.Receptor
getY() - Method in class NearOCR.Letter: Returns the current Y coordinate of this letter as it was once placed in the source document.
getY1() - Method in class NearOCR.Receptor
getY2() - Method in class NearOCR.Receptor

H

HyperbolicTangentThresholdModel - Class in NearNeural: Creates a new thresholding algorithm based on the hyperbolic tangent function.
HyperbolicTangentThresholdModel(double) - Constructor for class NearNeural.HyperbolicTangentThresholdModel: Creates a new tangental function with a given steepness.

I

image - Variable in class NearOCR.CharacterImage: The original image, in its original size
ImageFileFilter - Class in <Unnamed>: Filters files for use with a FileChooser, in order to allow selection of images only.
ImageFileFilter() - Constructor for class ImageFileFilter
IndexOutOfBoundsException - Exception in NearNeural: Represents an exception caused by attempting to address an absent Neuron or other Net element.
IndexOutOfBoundsException(String) - Constructor for exception NearNeural.IndexOutOfBoundsException: Creates a new exception with the message given.
IndexOutOfBoundsException() - Constructor for exception NearNeural.IndexOutOfBoundsException: Creates a new exception with no message
isPresent(Neuron) - Method in class NearNeural.NeuronLayer: checks if a Neuron is held in this layer.
isPresent(Edge) - Method in class NearNeural.ReceptiveNeuron: checks if an Edge is held in this layer.
ItemNotFoundException - Exception in NearNeural: Represents an exception caused by attempting to access an absent Neuron or other Net element.
ItemNotFoundException(String) - Constructor for exception NearNeural.ItemNotFoundException: Creates a new exception with the message given.
ItemNotFoundException() - Constructor for exception NearNeural.ItemNotFoundException: Creates a new exception with no message
itemStateChanged(ItemEvent) - Method in class TrainPanel: Fired when an itemevent firing component changes.

K

k - Variable in class NearNeural.WeightedThresholdModel: The coefficient value (steepness).

L

learningRate - Variable in class NearNeural.UnweightedBackPropagationMethod: The multiplier for error gradient descent.
Letter - Class in NearOCR: Stores a single character and forms the bulk of a Document.
Letter(BufferedImage) - Constructor for class NearOCR.Letter: Creates a new Letter from a given image, assuming scaled is identical.
LinearThresholdModel - Class in NearNeural: A threshold which is linear.This method may not work at all with some backpropagation methods
LinearThresholdModel(double) - Constructor for class NearNeural.LinearThresholdModel: Creates a new thresholding function with the given 'harshness'.
linkTo(NeuronLayer) - Method in class NearNeural.ReceptiveNeuronLayer: Links all neurons in this layer to all neurons in the layer provided.
LoadFrame - Class in <Unnamed>: Creates a new frame which allows the user to create a net with the parameters chosen.
LoadFrame(Frame, OCRManager) - Constructor for class LoadFrame: Creates and shows a new net building dialog.
loadImage(File) - Method in class NearOCR.OCRManager: Sets the current document's image source.
loadSymbols(File) - Method in class NearOCR.OCRManager: Loads a symbol table from the given file.
logError(String) - Static method in class NearOCR.Logger: Logs an error, sending it to all listeners and storing it internally.
Logger - Class in NearOCR: The logger is a client-agnostic way of reporting events and progress from the OCR manager to the interface.
Logger() - Constructor for class NearOCR.Logger
logInfo(String) - Static method in class NearOCR.Logger: Logs an information message, sending it to alll listeners and storing it internally.
logWarning(String) - Static method in class NearOCR.Logger: Logs a warning, sending it to all listeners and storing it internally.

M

main(String[]) - Static method in class OCRSwing: Runs the whole shebang.
MessageListener - Interface in NearOCR: An interface which allows people to receive global NearOCR events from the logger.
messageReceived(String) - Method in class ConsolePanel: Adds a new line to the log window when a logger event is fired.
messageReceived(String) - Method in interface NearOCR.MessageListener: Called when any form of message is logged.
moveNet(int, int) - Method in class NearOCR.NetStack: Moves a net in the list of nets.
moveNet(int, int) - Method in class NearOCR.OCRManager: Moves a net in the stack.

N

NearNeural - package NearNeural
NearOCR - package NearOCR
NetCell - Class in <Unnamed>: Shows a neural net as part of a JList.
NetCell() - Constructor for class NetCell: Creates a new net cell.
NetFileFilter - Class in <Unnamed>: Filters files for use with a FileChooser, in order to allow selection of neural nets only.
NetFileFilter() - Constructor for class NetFileFilter
NetPanel - Class in <Unnamed>: A panel, to be a tab, that represents net data to the user.
NetPanel(OCRManager) - Constructor for class NetPanel: Creates a new panel which displays net information.
NetStack - Class in NearOCR: The net stack holds all nets that are to be run durin the OCR process.
NetStack() - Constructor for class NearOCR.NetStack: Creates an empty net stack.
NeuralNet - Class in NearNeural: A practical implementation of a feed-forward back propogating multi-layer perceptron network.
NeuralNet(double[], int[], ThresholdingAlgorithm, BackPropagationMethod) - Constructor for class NearNeural.NeuralNet: Creates a new net with p_layers.length+1 layers, each with n nodes in them.
NeuralNet(FileReader, ThresholdingAlgorithm, BackPropagationMethod) - Constructor for class NearNeural.NeuralNet: Create a net from a saved net file.
NeuralNetException - Exception in NearNeural: A throwable exception that is the parent of all other exceptions regarding nets.
NeuralNetException(String) - Constructor for exception NearNeural.NeuralNetException: Creates a new NeuralNetException with the message given.
NeuralNetException() - Constructor for exception NearNeural.NeuralNetException: Creates a new exception with no message
NeuralVisualisation - Class in <Unnamed>: A panel that draws a visal representation of a neural net, displaying colour-coded edges and alpha-blended neurons.
NeuralVisualisation(NeuralNet) - Constructor for class NeuralVisualisation: Creates a new visualisation with the given net and dimension.
NeuralVisualisationFrame - Class in <Unnamed>: A panel that draws a visal representation of a neural net, displaying colour-coded edges and alpha-blended neurons.
NeuralVisualisationFrame(NeuralNet, int, int, String) - Constructor for class NeuralVisualisationFrame: Creates a new visualisation with the given net and dimension.
Neuron - Interface in NearNeural: Neurons are the staple of the net.
NeuronCell - Class in <Unnamed>: Displays the details of a single output neuron weight.
NeuronCell() - Constructor for class NeuronCell: Creates a new neuron cell with blank values
NeuronLayer - Class in NearNeural: A NeuronLayer keeps track of an arbitrary selection of neurons, each of which may have arbitrary or even circular links.
NeuronLayer(ThresholdingAlgorithm) - Constructor for class NearNeural.NeuronLayer: Creates a blank neuron layer with the threshold model provided.
neurons - Variable in class NearNeural.NeuronLayer: Stores all Neurons in this layer
NewNetFrame - Class in <Unnamed>: Creates a new frame which allows the user to create a net with the parameters chosen.
NewNetFrame(Frame, OCRManager) - Constructor for class NewNetFrame: Creates and shows a new net building dialog.

O

OCRManager - Class in NearOCR: Manages all OCR function.
OCRManager() - Constructor for class NearOCR.OCRManager: Creates a new OCR manager, with default original values.
OCRSwing - Class in <Unnamed>: Displays the main window of the OCR program and contains all other panels.
OCRSwing() - Constructor for class OCRSwing: A simple constructor to create the main window

P

paint(Graphics) - Method in class NeuralVisualisation: Draws the net based on values from edges which are stored in the reference vectors built during populateEdges and populateNeurons methods.
preCalculate() - Method in class NearNeural.NeuralNet: Precalculates all neuron values in order to speed up the net's calculation.
preCalculate() - Method in class NearNeural.ReceptiveNeuronLayer: Precalculates all neuron values based on their edges.
ProgressListener - Interface in NearOCR: Designed to allow objects to listen in on the progress of long OCR-related operations.
prune(double) - Method in class NearNeural.NeuralNet: Removes edges with absolute values below the threshold given.
prune(double) - Method in class NearNeural.ReceptiveNeuron: Removes Edges with an absolute weight under the threshold given.

R

RandomReceptorPattern - Class in NearOCR: A pattern of receptors with totally random position, orientation and length.
RandomReceptorPattern() - Constructor for class NearOCR.RandomReceptorPattern
ReceptiveNeuron - Class in NearNeural: A receptive Neuron is a form of Neuron that maintains Edges, which themselves point to other Neurons, be they static or themselves a form of Receptive Neuron.
ReceptiveNeuron(ThresholdingAlgorithm) - Constructor for class NearNeural.ReceptiveNeuron: Creates a new neuron with no edges.
ReceptiveNeuron(Edge[], ThresholdingAlgorithm) - Constructor for class NearNeural.ReceptiveNeuron: Creates a Neuron with Edges from an array.
ReceptiveNeuronLayer - Class in NearNeural: A Receptive form of the Neuron Layer, this class allows Neurons to maintain Edges to other Neurons.
ReceptiveNeuronLayer(ThresholdingAlgorithm) - Constructor for class NearNeural.ReceptiveNeuronLayer: Creates a new receptive neuron layer with the desired thresholding algorithm.
Receptor - Class in NearOCR: Represents one single-unit receptor.
Receptor(double, double, double, double) - Constructor for class NearOCR.Receptor
ReceptorManager - Class in NearOCR: Manages the currently loaded receptor net and its various properties.
ReceptorManager(ReceptorPattern, int) - Constructor for class NearOCR.ReceptorManager: Creates a new ReceptorManager with a given receptor pattern and number of receptors.
ReceptorManager() - Constructor for class NearOCR.ReceptorManager: Creates an empty set of receptors.
ReceptorManager(FileReader) - Constructor for class NearOCR.ReceptorManager: Loads a set of receptors from a file.
ReceptorPanel - Class in <Unnamed>: A panel that allows the user to regenerate and load receptors.
ReceptorPanel(OCRManager) - Constructor for class ReceptorPanel: Creates a new slice panel using the OCR object in question.
ReceptorPattern - Interface in NearOCR: An interface to allow creation of arbitrary receptor generating algorithms.
redraw() - Method in class ResultsPanel: Redraws the output pane with data read from the OCRManager as passed in the constructor.
RegularObliqueReceptorPattern - Class in NearOCR: A receptor pattern that is essentially single-hatching.
RegularObliqueReceptorPattern() - Constructor for class NearOCR.RegularObliqueReceptorPattern
remove(Receptor) - Method in class NearOCR.ReceptorManager: Removes a receptor from the set.
remove(int) - Method in class NearOCR.SymbolTable: Removes a symbol from the table by index.
removeEdge(Edge) - Method in class NearNeural.ReceptiveNeuron: Removes an edge from this neuron.
removeEdge(int) - Method in class NearNeural.ReceptiveNeuron: Removes an edge from the layer by index.
removeLayer() - Method in class NewNetFrame: Removes a provisional layer from the list
removeMessageListener(MessageListener) - Static method in class NearOCR.Logger: Removes a message listener such that it will no longer be sent events.
removeNet(NeuralNet) - Method in class NearOCR.NetStack: Removes a net from the internal list.
removeNet(int) - Method in class NearOCR.OCRManager: Removes a net from the internal list.
removeNet(NeuralNet) - Method in class NearOCR.OCRManager: Removes a net from the stack.
removeNeuron(Neuron) - Method in class NearNeural.NeuralNet: Removes a neuron and all edges that point to and from it.
removeNeuron(Neuron) - Method in class NearNeural.NeuronLayer: Removes a neuron from this layer.
removeNeuron(int) - Method in class NearNeural.NeuronLayer: Removes a neuron from the layer by index.
repopulate(ReceptorPattern, int) - Method in class NearOCR.ReceptorManager: Regenerates receptors using a set pattern.
repopulateList() - Method in class NetPanel: Repopulates the list from the OCR manager's net stack.
repopulateList() - Method in class NewNetFrame: Repopulates and redraws the net layer list.
repopulateList() - Method in class TrainPanel: Repopulates the list of neural nets
repopulateOutputsList() - Method in class SymbolPanel: Repopulates all weighted nodes form the current symbol
repopulateSymbolList() - Method in class SymbolPanel: Reloads and repopulates the list of symbols from the internally stored OCRManager.
resetWeights() - Method in class SymbolPanel: Resets all weights to 1
ResultsPanel - Class in <Unnamed>: Draws the results of running an OCR operation as text.
ResultsPanel(OCRManager) - Constructor for class ResultsPanel: Creates a new result reading panel linked to the given OCR manager.
run(NeuralNet[], CharacterImage, ProgressListener, ReceptorManager) - Static method in class NearOCR.RunManager: Runs a single image through all nets, returning the raw net output.
run() - Method in class RunDispatcher: Runs the analysis
run() - Method in class TrainingDispatcher: Runs the training analysis
run() - Method in class WeightAnalysisDispatcher: Runs the analyis
RunDispatcher - Class in <Unnamed>: Dispatches a run operation in a separate thread such that the UI remains responsive.
RunDispatcher(OCRManager, ProgressListener, double, double) - Constructor for class RunDispatcher: Creates a new rundispatcher and waits for the run() method to be called.
RunManager - Class in NearOCR: The RunManager is responsible for all OCR running operations which actually recognise characters rather than merely adjusting weight values.
RunManager() - Constructor for class NearOCR.RunManager
runOneImg(NeuralNet[], CharacterImage, SymbolTable, ProgressListener, double, double, ReceptorManager) - Static method in class NearOCR.RunManager: Loops through all of the nets, runs them and add it to a list of doubles, then returns the best fit symbol for those outputs.
RunPanel - Class in <Unnamed>: A panel that shows the progress of analysis.
RunPanel(OCRManager) - Constructor for class RunPanel: Creates a new run monitor panel, maintaining a reference to the OCR manager passed.
runSingleTrainingCycle(NeuralNet, Symbol, int, ProgressListener, int, double, ReceptorManager) - Static method in class NearOCR.TrainingManager: Runs a single training cycle for a single net and a single symbol.

S

save(FileWriter) - Method in class NearNeural.NeuralNet: Saves all net memory to a file.
save(FileWriter) - Method in class NearOCR.ReceptorManager: Saves the Receptor manager to the file provided.
save(FileWriter) - Method in class NearOCR.SymbolTable: Saves the suymbol table to the file provided.
save() - Method in class ResultsPanel: Starts a JFileChooser to save the output as stripped from the output pane.
setAllWeights(double) - Method in class NearOCR.OCRManager: Resets all symbol weights to the given number.
setBackPropagationMethod(BackPropagationMethod) - Method in class NearNeural.NeuralNet: Sets the BackPropagationMethod that willl be used to train the net.
setDelta(double) - Method in class NearNeural.ReceptiveNeuron: Sets the difference between current value and desired value during the backpropagation process.
setDocument(Document) - Method in class NearOCR.OCRManager: Sets the document to parse.
setEdgeWeight(int, double) - Method in class NearNeural.ReceptiveNeuron: Sets the weight of an edge, as indexed by the connection to the current Neuron.
setImage(BufferedImage) - Method in class NearOCR.CharacterImage: Sets the original image of this Symbol from which the scaled one is calculated.
setImage(BufferedImage) - Method in class NearOCR.Document: Sets the image that this document is based on.
setIndex(int) - Method in class NearOCR.WeightedNode: Sets the index number of this node.
setInput(int) - Method in class NearOCR.NetStack: Attempts to adjust the number of input neurons.
setInputValue(double, int) - Method in class NearNeural.NeuralNet: Allows the user to set the value of one input Neuron.
setInputValue(double, int) - Method in class NearNeural.StaticNeuronLayer: Sets the input value of one static node.
setInputValues(double[]) - Method in class NearNeural.NeuralNet: Sets input values from an array of doubles.
setInputValues(double[]) - Method in class NearNeural.StaticNeuronLayer: Sets all nodes in this layer to the values given, in order.
setK(double) - Method in class NearNeural.WeightedThresholdModel: Sets the coefficient of the running function (steepness).
setLearningRate(double) - Method in class NearNeural.UnweightedBackPropagationMethod: Sets the learning rate of this backpropagation method.
setLength(int) - Method in class NearOCR.Symbol: Sets the number of output nodes, which ought to be equal to the number of total output nodes in the OCR manager.
setLength(int) - Method in class NearOCR.SymbolTable: Sets the number of weighted nodes every symbol in the table has.
setOldWeight(double) - Method in class NearNeural.Edge: Set the old weight of this Edge.
setReceptorManager(ReceptorManager) - Method in class NearOCR.OCRManager: Sets the receptor manager.
setTarget(double) - Method in class NearOCR.WeightedNode: Sets the target value, from which error is calculated
setValue(double) - Method in class NearNeural.StaticNeuron: Sets the value of this neuron.
setValue(String) - Method in class NearOCR.CharacterImage: Sets the string value of this Symbol.
setWeight(double) - Method in class NearNeural.Edge: Sets the weight of this edge
setWeight(WeightedNode, int) - Method in class NearOCR.Symbol: Sets the weight of one particular output node, indexed by value in internal array - NOT by the index as stored in the weighted node itself.
setWeight(double) - Method in class NearOCR.WeightedNode: Sets the weight of this node, by which the output of the net will be multiplied in order to fin its true output.
setX(int) - Method in class NearOCR.Letter: Sets the X co-ordinate of the letter on the page.
setX1(double) - Method in class NearOCR.Receptor
setX2(double) - Method in class NearOCR.Receptor
setY(int) - Method in class NearOCR.Letter: Sets the Y coordinate of the letter as placed in the source document.
setY1(double) - Method in class NearOCR.Receptor
setY2(double) - Method in class NearOCR.Receptor
SigmoidThresholdModel - Class in NearNeural: Represents a thresholding algorithm based on the sigmoid function.
SigmoidThresholdModel(double) - Constructor for class NearNeural.SigmoidThresholdModel: Creates a new thresholding function with the given 'harshness'.
SinusoidalThresholdModel - Class in NearNeural: Creates a new sinusoidal thresholding algorithm.
SinusoidalThresholdModel(double) - Constructor for class NearNeural.SinusoidalThresholdModel: Creates a new sinusoidal function with a given steepness.
slice(double, double) - Method in class NearOCR.OCRManager: Attempts to find letters by dividing an image up by lines and letters.
slice() - Method in class SlicePanel: Reslices all images according to slider values, causing the OCR manager to update its own letters.
SlicePanel - Class in <Unnamed>: A panel that allows the user to choose how to slice images up.
SlicePanel(OCRManager) - Constructor for class SlicePanel: Creates a new slice panel using the OCR object in question.
source - Variable in class NearNeural.Edge: Stores a reference to the source neuron.
stateChanged(ChangeEvent) - Method in class SlicePanel: Handles a value change in the threshold sliders, recalculating slices each time the slider is dropped.
StaticNeuron - Class in NearNeural: Represents an input neuron, which has a static value, uninfluenced by any edges.
StaticNeuron(double, ThresholdingAlgorithm) - Constructor for class NearNeural.StaticNeuron: Creates a Neuron with the value given.
StaticNeuronLayer - Class in NearNeural: The Static Neuron Layer holds only Static neurons, that is those without any Edges.
StaticNeuronLayer(int, ThresholdingAlgorithm) - Constructor for class NearNeural.StaticNeuronLayer: Creates a new layer with a given number of Static Neurons.
StaticNeuronLayer(ThresholdingAlgorithm) - Constructor for class NearNeural.StaticNeuronLayer
supportedVersions - Variable in class NearNeural.NeuralNet: The versions from which this version can load files.
Symbol - Class in NearOCR: A symbol is responsible for managing training and finding values from net output.
Symbol(String, String) - Constructor for class NearOCR.Symbol: Creates a new symbol with a given value and from a given filename.
SymbolCell - Class in <Unnamed>: Shows a symbol as part of a JList.
SymbolCell() - Constructor for class SymbolCell: Creates a new symbol cell.
SymbolFileFilter - Class in <Unnamed>: Filters files for use with a FileChooser, in order to allow selection of images only.
SymbolFileFilter() - Constructor for class SymbolFileFilter
SymbolPanel - Class in <Unnamed>
SymbolPanel(OCRManager) - Constructor for class SymbolPanel: Creates a new symbol editing panel linked to the given OCRManager.
SymbolTable - Class in NearOCR: Stores a table of all outputs correlating to their ideal output as a character and the filepath of the original image which was used to train the net, along with an average 'off' weighting.
SymbolTable() - Constructor for class NearOCR.SymbolTable: Creates a blank symbol table
SymbolTable(FileReader) - Constructor for class NearOCR.SymbolTable: Creates a new symbol table from a saved symbol table file.

T

ThreadedUnweightedBackPropagationMethod - Class in NearNeural: A threaded version of UnweightedBackPropagationMethod which uses a standard gradient-descent backpropagation method.
ThreadedUnweightedBackPropagationMethod(double) - Constructor for class NearNeural.ThreadedUnweightedBackPropagationMethod: Creates a new ThreadedUnweightedBackPropagationMethod with the given learning rate.
ThreePointRuleDifferentiation - Class in NearNeural: Implements a numerical method for differentiation.
ThreePointRuleDifferentiation(double) - Constructor for class NearNeural.ThreePointRuleDifferentiation: Creates a new differentiation constant necessary to approximate differential.
ThresholdingAlgorithm - Interface in NearNeural: A formal definition of an algorithm which adjusts input values.
thresholdModel - Variable in class NearNeural.NeuronLayer: The thresholding model to use for back propagation
toString() - Method in class NearNeural.NeuralNet: Returns a string representing the object.
toString() - Method in class NearOCR.Receptor
toString() - Method in class NearOCR.Symbol: Returns this Symbol as represented by a string.
toString() - Method in class NearOCR.WeightedNode: Returns this node represented as a string.
train(double[]) - Method in class NearNeural.NeuralNet: Runs the back propogation algorithms on the whole net.
train() - Method in class TrainPanel: Trains the net using the values as in the ui
trainAllSymbols(NeuralNet, Symbol[], int, int, ProgressListener, double, int, double, ReceptorManager) - Static method in class NearOCR.TrainingManager: Trains the net in a cyclic fashion with all symbols given.
TrainingDispatcher - Class in <Unnamed>: Dispatches training threads, allowing the UI to continue updating.
TrainingDispatcher(NeuralNet, Symbol[], int, int, TrainPanel, double, int, double, OCRManager) - Constructor for class TrainingDispatcher: Creates a new training dispatcher with the parmeters given.
TrainingManager - Class in NearOCR: Handles all aspects relating to training techniques for both the encapsulated Neural Net and the symbol table.
TrainingManager() - Constructor for class NearOCR.TrainingManager
TrainPanel - Class in <Unnamed>: A panel that allows the user to train nets in the net stack.
TrainPanel(OCRManager) - Constructor for class TrainPanel: Creates a new panel which displays net information.

U

UnresolvedValueException - Exception in NearNeural: An exception that handles values which are not processable as to the user's requests.
UnresolvedValueException(String) - Constructor for exception NearNeural.UnresolvedValueException: Creates a new UnresolvedValueException with the message given.
UnresolvedValueException() - Constructor for exception NearNeural.UnresolvedValueException: Creates a new exception with no message
UnweightedBackPropagationMethod - Class in NearNeural: Backpropagates a layer without any kind of weighting to avoid local minima of error.
UnweightedBackPropagationMethod(double) - Constructor for class NearNeural.UnweightedBackPropagationMethod: Creates a new backpropagation object with the given learning rate.
update() - Method in class NeuralVisualisation: Recalculates edge and neurons lists from the net.
update() - Method in class NeuralVisualisationFrame

V

validateInput(double) - Method in class NearNeural.HyperbolicTangentThresholdModel: Validates input against the applicable range for this thresholding algorithm.
validateInput(double) - Method in class NearNeural.LinearThresholdModel: Validates input against the applicable range for this thresholding algorithm.
validateInput(double) - Method in class NearNeural.SigmoidThresholdModel: Validates input values.
validateInput(double) - Method in class NearNeural.SinusoidalThresholdModel: Validates input against the applicable range for this thresholding algorithm.
validateInput(double) - Method in class NearNeural.ThreePointRuleDifferentiation: Validates input against the applicable range for this thresholding algorithm.
validateInput(double) - Method in interface NearNeural.ThresholdingAlgorithm: Returns true if the provided number is within reasonable input bounds for this algorithm.
value(Edge[]) - Method in class NearNeural.HyperbolicTangentThresholdModel: Calculates Neuron value from all Edges and their source Neurons.
value(Edge[]) - Method in class NearNeural.LinearThresholdModel: Calculates Neuron value from all Edges and their source Neurons.
value() - Method in interface NearNeural.Neuron: Returns the value of this Neuron, calculated through any means.
value() - Method in class NearNeural.ReceptiveNeuron: Returns the sum of all Edge values, multiplied by their weight.
value(Edge[]) - Method in class NearNeural.SigmoidThresholdModel: Calculates Neuron value from all Edges and their source Neurons.
value(Edge[]) - Method in class NearNeural.SinusoidalThresholdModel: Calculates Neuron value from all Edges and their source Neurons.
value() - Method in class NearNeural.StaticNeuron: Returns this Static Neuron's value.
value(Edge[]) - Method in class NearNeural.ThreePointRuleDifferentiation: Calculates Neuron value from all Edges and their source Neurons.
value(Edge[]) - Method in interface NearNeural.ThresholdingAlgorithm: Returns the value of a neuron.
value - Variable in class NearOCR.CharacterImage: Stores this symbol's value
valueChanged(double, double) - Method in interface NearOCR.ProgressListener: Called when the progress is updated.
valueChanged(ListSelectionEvent) - Method in class NetPanel: Is called whenever the value of a list changes, in this case this is the Net list only.
valueChanged(double, double) - Method in class RunPanel: Fired when a value is changed in the analysis process.
valueChanged(ListSelectionEvent) - Method in class SymbolPanel: Fires when someone makes a selection on either list.
valueChanged(double, double) - Method in class SymbolPanel: Fired when a value is changed in the analysis process.
valueChanged(ListSelectionEvent) - Method in class TrainPanel: Changed when the list selection changes.
valueChanged(double, double) - Method in class TrainPanel: Fired when a value is changed in the training process.
ValueOutOfBoundsException - Exception in NearNeural: An exception that is generally thrown when a value is out of acceptable bounds.
ValueOutOfBoundsException(String, double) - Constructor for exception NearNeural.ValueOutOfBoundsException: Creates a new ValueOutOfBoundsException with the message and value given.
ValueOutOfBoundsException(String) - Constructor for exception NearNeural.ValueOutOfBoundsException: Creates a new ValueOutOfBoundsException with the given message.
ValueOutOfBoundsException() - Constructor for exception NearNeural.ValueOutOfBoundsException: Creates a new exception with no message
version - Variable in class NearNeural.NeuralNet: The version of the net code.

W

WeightAnalysisDispatcher - Class in <Unnamed>: Dispatches weight analysis to adjust all symbol weighted nodes ot the current net stack.
WeightAnalysisDispatcher(OCRManager, ProgressListener) - Constructor for class WeightAnalysisDispatcher: Creates a new dispatcher with the given values, which then waits to be run().
WeightedNode - Class in NearOCR: Holds an adjusted weight for one output.
WeightedNode(int, double, double) - Constructor for class NearOCR.WeightedNode: Creates a new weighted node with a set index, weight and target.
WeightedThresholdModel - Class in NearNeural: This class contains the necessary code to implement coefficient management in order to adjust a thresholding function.
WeightedThresholdModel(double) - Constructor for class NearNeural.WeightedThresholdModel: Cannot be called directly: allows subclasses to create weighted copies.

A B C D E F G H I K L M N O P R S T U V W