摘要:
A learning and recognition machine has a major classification section for classifying an input pattern into a plurality of category groups and calculating a group belongingness to each category group and a plurality of fine classification sections for calculating a similarity within each category group. The machine causes the plurality of fine classification sections to learn in a coordinated manner by reflecting the group belongingness of the input pattern to update of a weight factor implemented in the fine classification sections, thereby allowing a recognition rate of a pattern situated at the boundary of the category groups to be improved and an effective supplemental learning to be performed.
摘要:
A learning type signal recording and reproducing apparatus has a signal detector for detecting signals recorded on recording media, a teaching signal generator for generating reference teaching signals having predetermined waveform signal patterns, and a learning type waveform converter for generating converting rules from signals recorded on the recording medium to original signals corresponding to the recorded signals, by learning the relationship between the detection results by the signal detector of teaching signals recorded on the recording media and reference teaching signals outputted by the teaching signal generator. With this apparatus, signals can be stably reproduced even if the signals contain distortion caused by noise or difference in the recording media.
摘要:
The signal recording and reproducing apparatus of this invention includes: a signal detector for detecting a teacher signal and an information signal recorded on a recording medium; a converting section for converting the teacher signal and the information signal detected by the signal detector into a reproduction teacher signal and a reproduction information signal, respectively, based on a predetermined conversion rule; and a teacher signal generator for generating a reference teacher signal, wherein the teacher signal is first detected before the detection of the information signal, and the converting means includes a learning type waveform converting section for automatically establishing the predetermined conversion rule based on the reproduction teacher signal and the reference teacher signal.
摘要:
The invention relates to an interface apparatus for making input and output of appliances having display such as computer, word processor, information appliance and television, comprising recognizing means for recognizing the shape or move of the hand of an operator, display means for displaying the features of the shape or move of the hand recognized by the recognizing means as special shape in the screen, and control means for controlling the information displayed in the screen by the special shape displayed in the screen by the display means, wherein the two-dimensional or three-dimensional information displayed in the screen can be selected, indicated or moved only by changing the shape or moving the hand, so that the interface apparatus of very excellent controllability and high diversity may be presented.
摘要:
The present invention relates to an image processing apparatus in which gradation correction on each partial region is performed between an image input means and an edge extracting means so that an edge of a region within an image is accurately extracted. An image dividing means 2, which had already learned a relationship between the condition of an image and the necessity of image division so as to judge the condition of the image based on learning results and divide the image appropriately, divides an input image from an image input means 1 into a plurality of image blocks. An image correction information extracting means 3 calculates correction information on each image block. A gradation correction means 4, which had already learned a relationship between correction information and non-linear correction curves to judge correction information based on learning results and perform gradation correction on a target image block using a selected curve, performs gradation correction and enhances an edge which is to be extracted. An image synthesizing means 5 recombines the respective corrected image blocks. An edge extracting means 6 extracts an edge from a synthesized image. An image output means 7 outputs an edge image.
摘要:
In a multi-layered neural network circuit provided with an input layer having input vectors, an intermediate layer having networks in tree-like structure whose outputs are necessarily determined by the values of the input vectors and whose number corresponds to the number of the input vectors of the input layer, and an output layer having plural output units for integrating all outputs of the intermediate layer, provided are learning-time memories for memorizing the numbers of times at learning in paths between the intermediate layer and the respective output units, threshold processing circuits for threshold-processing the outputs of the leaning-time memories, and connection control circuits to be controlled by the outputs of the threshold processing circuits for controlling connection of paths between the intermediate layer and the output units. The outputs of the intermediate layer connected by the connection control circuits are summed in each output unit. Thus, the neural network circuit for recognizing an image or the like can execute recognition and learning of data to be recognized at high speed with small circuit size, and the recognition accuracy for unlearned data is high.
摘要:
There is disclosed a character recognition machine adapted to recognize Japanese characters such as kanjis and kanas. The machine comprises a character string storage portion, a character extraction portion, a character recognition portion, and a language processing portion. A character string to be recognized is stored as an image in the storage portion. The character extraction portion comprises a network consisting a plurality of interconnected operators each of which has numerous inputs and outputs. An evaluation function which assumes its minimum value when a character extraction produces the best results is calculated by the operators simultaneously so as to minimize the value of the function. The character recognition portion calculates degrees of similarity of a character pattern to various character categories, the character pattern being applied from the character extraction portion. The language processing portion receives these degrees of similarity and selects a character category which seems to provide the most correct combination of characters in terms of vocabulary and grammar.
摘要:
There is disclosed a character recognition machine adapted to recognize Japanese characters such as kanjis and kanas. The machine comprises a character string storage portion, a character extraction portion, a character recognition portion, and a language processing portion. A character string to be recognized is stored as an image in the storage portion. The character extraction portion comprises a network consisting a plurality of interconnected operators each of which has numerous inputs and outputs. An evaluation function which assumes its minimum value when a character extraction produces the best results is calculated by the operators simultaneously so as to minimize the value of the function. The character recognition portion calculates degrees of similarity of a character pattern to various character categories, the character pattern being applied from the character extraction portion. The language processing portion receives these degrees of similarity and selects a character category which seems to provide the must correct combination of characters in terms of vocabulary and grammar.
摘要:
In an information processing apparatus for implementing a neural network, if an input vector is inputted to a calculating unit, a neuron which responds to the input vector is retrieved in accordance with network interconnection information stored in a first storage unit and the neuron number indicating the retrieved neuron is written in a first register. The calculating unit reads out the internal information of the neuron stored in a second storage unit by using the neuron number, writes it in a second register, and calculates the sum of products of the outputs of the neurons and the connection loads of synapses connected to the neurons. By repeating the sequence of operations by the number of times corresponding to the total number of input vectors, a recognition process is executed. The neural network can easily be expanded by rewriting the contents of the first and second storage units.
摘要:
A recognizing and judging apparatus has a network organized in a multilayered hierarchical manner and a plurality of branched tree structures corresponding to the number of inputted data. The branched tree structures are organized by a plurality of recognition units, each of which includes a signal input section and a quantizer for performing a quantization according to a signal inputted from the signal input section. Each of the recognition units further includes a path input section having at least one path input terminal, a path output section having at least one path output terminal, and a path selecting section operatively coupled with both the path input section and the path output section for performing a selection of paths according to an output of the quantizer.