Patent US5544279A successfully refreshed from Google Patents.
US5544279A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Fuzzy logic recognition device which determines the mode of an unknown move using fuzzy set theory
Assignee: United Microelectronics Corp
•
Filing: 1994-10-13
•
Granted: 1996-08-06
•
Expires: 2014-10-13
Abstract
Abstract A fuzzy logic recognition device has a plurality of sets of feature extraction windows, each set storing the predetermined features for one feature class therein, and comparing the stored features with the corresponding features of an unknown recognition mode to output a logic "1" via one corresponding feature extraction window where an identicalness condition occurs, and logic "0" via the other feature extraction windows where no identicalness condition occurs. A plurality of sets of membership function generators are coupled to the feature extraction window sets respectively. Each membership function generator which receives the logic "1" generates the memberships of the corresponding feature with respect to the target modes, and the membership function generators which receive the logic "0" do not generate the memberships. A minimum gate is coupled to the membership function generator sets for comparing the memberships with respect to the same target mode, respectively, and for outputting the smallest membership for every one of the target modes. A maximum gate is coupled to the minimum gate for comparing the memberships outputted from the minimum gate means, and for outputting the largest membership to indicate that the unknown recognition mode is which one of the target mode.
Inventors
- 👤 Zhi-Jian Li
- 👤 Bing-Xue Shi
- 👤 Bin-Qiao Li
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 22:31
CPC Classifications
G; G06; G06V; G06V10/00; G06V10/70; G06V10/74; G06V10/75
Claims Analysis
Total Claims: 11
Claim 1
Independent Claim
272 words
A fuzzy logic recognition device for recognition of an unknown recognition mode as one of a plurality of target modes, wherein a plurality of predetermined feature classes are selected for the unknown recognition and target modes, and a plurality of predetermined features are selected for each feature class for the target modes, each predetermined feature having a membership with respect to each target mode, said fuzzy logic recognition device comprising: a plurality of sets of feature extraction window means each for storing the predetermined features for one feature class therein, for comparing the stored features with the corresponding features of the unknown recognition mode, for outputting a positive binary signal via one corresponding feature extraction window means where an identicalness condition occurs, and for outputting negative binary signals via the other feature extraction window means where no identicalness condition occurs; a plurality of sets of membership function generators for receiving respective outputs of said feature extraction window means, each membership function generator that receives said positive binary signal generating the memberships of the corresponding predetermined feature with respect to the target modes, and the membership function generators that receive said negative binary signals not generating the memberships; a minimum gate means, coupled to said membership function generator sets, for comparing the memberships with respect to the same target mode, respectively, and for outputting the smallest membership for every one of the target modes; and a maximum gate means, coupled to said minimum gate means, for comparing said memberships outputted from said minimum gate means, and for outputting the largest membership to indicate the unknown recognition mode as one of the target modes.
Claim 2
Dependent Claim
272 words
A fuzzy logic recognition device for recognition of an unknown recognition mode as one of a plurality of target modes, wherein a plurality of predetermined feature classes are selected for the unknown recognition and target modes, and a plurality of predetermined features are selected for each feature class for the target modes, each predetermined feature having a membership with respect to each target mode, said fuzzy logic recognition device comprising: a plurality of sets of feature extraction window means each for storing the predetermined features for one feature class therein, for comparing the stored features with the corresponding features of the unknown recognition mode, for outputting a positive binary signal via one corresponding feature extraction window means where an identicalness condition occurs, and for outputting negative binary signals via the other feature extraction window means where no identicalness condition occurs; a plurality of sets of membership function generators for receiving respective outputs of said feature extraction window means, each membership function generator that receives said positive binary signal generating the memberships of the corresponding predetermined feature with respect to the target modes, and the membership function generators that receive said negative binary signals not generating the memberships; a minimum gate means, coupled to said membership function generator sets, for comparing the memberships with respect to the same target mode, respectively, and for outputting the smallest membership for every one of the target modes; and a maximum gate means, coupled to said minimum gate means, for comparing said memberships outputted from said minimum gate means, and for outputting the largest membership to indicate the unknown recognition mode as one of the target modes.
Claim 3
Dependent Claim
106 words
The fuzzy logic recognition device as claimed in claim 1, further comprising a plurality of NOR gates, and a plurality of unit feature membership function generators connected respectively between said NOR gates and said minimum gate, each NOR gate being connected to the outputs of one corresponding feature extraction window means set at its input terminals to generate a logic high signal when the outputs of said one feature extraction window means set are all said negative binary signal, and then said logic high signal activating one corresponding unit feature membership function generator to generate a plurality of unit memberships with respect to the target modes.
Claim 4
Dependent Claim
158 words
The fuzzy logic recognition device as claimed in claim 2, wherein each of said membership function generators is implemented by a membership function generating circuit including a PMOS model transistor, a plurality of PMOS mirror transistors, and an NMOS control transistor with its gate electrode controlled by the output of the corresponding feature extraction window means, its source electrode connected to ground, and its drain electrode connected to the gate and drain electrodes of said model transistor, said model and mirror transistors being connected in parallel with their source electrodes connected to a power source and their gate electrodes connected together, and the channel lengths and the channel widths of said mirror transistors being respectively adjusted with respect to the channel length and the channel width of said model transistor, so that the mirror currents presented at the drain electrodes of said mirror transistors can represent the memberships of the corresponding feature with respect to the target modes.
Claim 5
Dependent Claim
39 words
The fuzzy logic recognition device as claimed in claim 3, further comprising a current-to-voltage converter coupled to said maximum gate means in order to convert the current mode of the output of said maximum gate into a voltage mode.
Claim 6
Dependent Claim
19 words
The fuzzy logic recognition device as claimed in claim 4, wherein said predetermined features are represented in binary form.
Claim 7
Dependent Claim
61 words
The fuzzy logic recognition device as claimed in claim 5, further comprising means, coupled to said maximum gate means, for receiving a threshold current to be compared with said memberships in said maximum gate means, and said maximum gate means outputting logic "0" signals to indicate that the recognition is rejected when said memberships are all smaller than said threshold current.
Claim 8
Dependent Claim
139 words
The fuzzy logic recognition device as claimed in claim 2, wherein each of said unit feature membership function generators is implemented by a membership function generating circuit including a PMOS model transistor, a plurality of PMOS mirror transistors, and an NMOS control transistor with its gate electrode controlled by the output of the corresponding NOR gate, its source electrode connected to ground, and its drain electrode connected to the gate and drain electrodes of said model transistor, said model and mirror transistors being connected in parallel with their source electrodes connected to a power source and their gate electrodes connected together, and said mirror transistors having the same channel length and the channel width, so that the mirror currents presented at the drain electrodes of said mirror transistors can represent the unit memberships with respect to the target modes.
Claim 9
Dependent Claim
39 words
The fuzzy logic recognition device as claimed in claim 7, further comprising a current-to-voltage converter coupled to said maximum gate means in order to convert the current mode of the output of said maximum gate into a voltage mode.
Claim 10
Dependent Claim
19 words
The fuzzy logic recognition device as claimed in claim 8, wherein said predetermined features are represented in binary form.
Claim 11
Dependent Claim
61 words
The fuzzy logic recognition device as claimed in claim 9, further comprising means, coupled to said maximum gate means, for receiving a threshold current to be compared with said memberships in said maximum gate means, and said maximum gate means outputting logic "0" signals to indicate that the recognition is rejected when said memberships are all smaller than said threshold current.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1999-08-06 | 2000-02-05 | 2000-08-06 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2003-08-06 | 2004-02-05 | 2004-08-06 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2007-08-06 | 2008-02-05 | 2008-08-06 | $7700.00 | Window Closed / Lapsed |