| 摘要 |
The invention consists of e-circuits built from basic modules of e-cells (271, 272 and 273) which are capable of recognizing a previously memorized percept anywhere within an arbitrarily large input field without incurring delay related to the size of the search space. The invention is also capable of isolating a previously memorized percept (278) within the input field when in the adjacent presence of other percepts and closely related distractors, locating and recognizing all occurrences of a repeated subfield within the input field, even in the presence of closely related distractors and performing sequential shift attention between a number of different percepts. The invention is applicable to various recognition tasks including speech and music recognition, vision, olefaction and touch. The invention also consists of a method to apply these e-circuits to the task of learning and recognizing subjects in images under translation, and if necessary scaling, rotation and other transforms.
|
