System for identifying information represented by biological signals
This system for identifying information represented by biological signals is configured so as to detect biological signals (S 501 ), analyze the detected biological signals and then output feature data (S 502 ), determine the respective similarities between the feature data and a plurality of teaching data (S 503 ), store the similarities per time in a time series (S 504 ), and determine information represented by the biological signals on the basis of the plurality of similarities within a prescribed period among the stored similarities in the time series (S 505 ).
1. A computer apparatus for identifying information represented by a biological signal, the computer apparatus comprising:
receiver means for receiving a biological signal;
analysis means for analyzing the received biological signal and outputting feature data;
first determination means for determining output vectors indicating degrees of similarity between the feature data and each of a plurality of teaching data;
storage means for chronologically storing the output vectors for each time; and
second determination means for determining information represented by the biological signal based on a plurality of output vectors within a predetermined period in the chronological output vectors stored in the storage means, wherein the second determination means:
calculates computed values for each one of the plurality of teaching data by performing a computation on degrees of similarity indicated by each of corresponding chronological components of the plurality of output vectors within the predetermined period, wherein each of the degree of similarity on which the computation is performed has a value in a range of 0 to 1, wherein an end point of the predetermined period is a last time at which an output vector is stored in the storage means; and
determines the information represented by the biological signal based on the computed values.
2. The computer apparatus of claim 1 , wherein the second determination means extracts teaching data corresponding to a highest computed value among the computed values and determines information indicated by the extracted teaching data as the information represented by the biological signal.
3. The computer apparatus of claim 1 , wherein the second determination means extracts at least one teaching data corresponding to a computed value exceeding a predetermined threshold value among the computed values and determines information indicated by the extracted teaching data as the information represented by the biological signal.
4. The computer apparatus of claim 3 , wherein the second determination means extracts a plurality of teaching data corresponding to computed values exceeding the predetermined threshold value among the computed values and determines information indicated by each of the plurality of extracted teaching data as the information represented by the biological signal.
5. The computer apparatus of claim 4 , wherein the information represented by the biological signal indicates that a composite motion has been performed.
6. The computer apparatus of claim 1 , wherein the computed values are total values.
7. The computer apparatus of claim 1 , wherein the storage means is a buffer for temporarily storing information, and the output vectors are temporarily stored in the buffer.
8. The computer apparatus of claim 1 , wherein the predetermined period is about 10 to 200 ms.
9. The computer apparatus of claim 1 for finger rehabilitation, for swallow diagnosis, for a wheelchair, for a prosthetic hand, for a prosthetic arm, for a prosthetic foot, for a robot, for an upper limb assisting apparatus, for a lower limb assisting apparatus, or for a trunk assisting apparatus.
10. A method for identifying information represented by a biological signal, the method comprising:
receiving a biological signal;
analyzing the received biological signal and outputting feature data;
determining output vectors indicating degrees of similarity between the feature data and each of a plurality of teaching data;
chronologically storing the output vectors for each time in a storage means; and
determining information represented by the biological signal based on a plurality of output vectors within a predetermined period in the chronological output vectors stored in the storage means, wherein determining information represented by the biological signal comprising:
calculating computed values for each one of the plurality of teaching data by performing a computation on degrees of similarity indicated by each of corresponding chronological components of the plurality of output vectors within the predetermined period, wherein each of the degrees of similarity on which the computation is performed has a value in a range of 0 to 1, wherein an end point of the predetermined period is a last time at which an output vector is stored in the storage means; and
determining the information represented by the biological signal based on the computed values.
11. A non-transitory storage medium storing a program for identifying information represented by a biological signal, the program is executed in a computer apparatus comprising a processor unit and a memory unit, the program cause the processor unit to perform a processing comprising:
receiving a biological signal;
analyzing the received biological signal and outputting feature data;
determining output vectors indicating degrees of similarity between the feature data and each of a plurality of teaching data;
chronologically storing the output vectors for each time in the memory unit; and
determining information represented by the biological signal based on a plurality of output vectors within a predetermined period in the chronological output vectors stored in the memory unit, wherein determining information represented by the biological signal comprising:
calculating computed values for each one of the plurality of teaching data by performing a computation on degrees of similarity indicated by each of corresponding chronological components of the plurality of output vectors within the predetermined period, wherein each of the degrees of similarity on which the computation is performed has a value in the range of 0 to 1, wherein an end point of the predetermined period is a last time at which an output vector is stored in the memory unit; and
determining the information represented by the biological signal based on the computed values.