Neural network based pattern recognizer
원문보기
IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0344344
(2008-12-26)
|
등록번호 |
US-8229209
(2012-07-24)
|
발명자
/ 주소 |
|
출원인 / 주소 |
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
1 인용 특허 :
2 |
초록
▼
An image-based pattern recognizer and a method and apparatus for making such a pattern recognizer are disclosed. By employing positional coding, the meaning of any feature present in an image can be defined implicitly in space. The pattern recognizer can be a neural network including a plurality of
An image-based pattern recognizer and a method and apparatus for making such a pattern recognizer are disclosed. By employing positional coding, the meaning of any feature present in an image can be defined implicitly in space. The pattern recognizer can be a neural network including a plurality of stages of observers. The observers are configured to cooperate to identify the presence of features in the input image and to recognize a pattern in the input image based on the features. Each of the observers includes a plurality of neurons. The input image includes a plurality of units, and each of the observers is configured to generate a separate output set that includes zero or more coordinates of such units.
대표청구항
▼
1. A processing system comprising: a processor; anda memory coupled to the processor and storing code which, when executed by the processor, causes the processing system to perform a process of using a neural network to process an image, the process including applying an input image to the neural ne
1. A processing system comprising: a processor; anda memory coupled to the processor and storing code which, when executed by the processor, causes the processing system to perform a process of using a neural network to process an image, the process including applying an input image to the neural network, the input image including a plurality of units, the neural network including a plurality of observers, each observer including a plurality of neurons, each neuron configured to execute an integrate-and-fire process, each observer configured to generate an output that includes zero or more coordinates; andrecognizing a pattern in the input image by processing the input image with the neural network, wherein firing of any particular neuron indicates the presence of a category of patterns in the input image, at a position corresponding to the particular neuron, wherein processing the input image with the neural network includes,for each coordinate in an output of each of the plurality of observers, integrating a corresponding potential over a range of time slices, andfor each said unit whose potential exceeds a threshold after integration by the observer, including the coordinate of the unit in an output set of the observer. 2. A processing system as recited in claim 1, wherein an output of each neuron is processed by a downstream observer only if the output of the neuron represents a detected event. 3. A processing system as recited in claim 1, further comprising: automatically triggering a specified action in response to recognizing the pattern in the input image. 4. A method comprising: applying an input image to a neural network in a machine-implemented processing system, the neural network including a plurality of observers, the input image including a plurality of units, each observer including a plurality of neurons, each neuron configured to execute an integrate-and-fire process, wherein each of the observers generates an output that includes zero or more coordinates; andrecognizing a pattern in the input image in the machine-implemented processing system by processing the input image with the neural network, wherein firing of any particular neuron indicates the presence of a category of patterns in the input image, at a position corresponding to the particular neuron, and wherein processing the input image with the neural network includes, for each coordinate in an output of each of the plurality of observers, integrating a corresponding potential over a range of time slices, andfor each said unit whose potential exceeds a threshold after integration by the observer, including the coordinate of the unit in an output set of the observer. 5. A method as recited in claim 4, wherein an output of each neuron is processed by a downstream observer only if the output of the neuron represents a detected event. 6. A method as recited in claim 4, further comprising: automatically triggering a specified action in response to recognizing the pattern in the input image.
이 특허에 인용된 특허 (2)
-
Paquier, Williams J. F., Multi-stage image pattern recognizer.
-
Huang Hsin-Hao (Kaohsiung TWX) Lin Shui-Shun (Tallahassee FL) Knapp Gerald M. (Baton Rouge LA) Wang Hsu-Pin (Tallahassee FL), Supervised training of a neural network.
이 특허를 인용한 특허 (1)
-
Bian, Qiao-Ling; Lu, Yan-Qing; Wang, Jin, Image processing method for removing moving object and electronic device.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.