General purpose recognition e-circuits capable of translation-tolerant recognition, scene segmentation and attention shift, and their application to machine vision
The invention comprises e-circuits built from basic modules of e-cells which are capable of: recognizing a previously memorized percept anywhere within an arbitrarily large input field without incurring delay related to size of the search space; isolating a previously memorized percept 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 in the input field. The invention is applicable to various recognition tasks including those in sensory domains such as speech and music recognition, vision, olefaction, and touch. The invention also comprises a method to apply these e-circuits specifically to the task of learning and recognizing subjects in images in any two dimensional signal space under translation, and if necessary, scaling, rotation and other transforms.
This application is related to provisional application number 60/067,125 Title: "Methods and Apparatuses For Implementing Equivalent Cells, Memory Circuits, General Purpose Recognition Circuits, and Machine Vision Systems, filed Dec. 2, 1997." It is also related to utility patent application, serial number 09/201,395, entitled "E-Cell (Equivalent Cell) and The Basic Circuit Modules of E-Circuits: E-Cell Pair Totem, The Basic Memory Circuit and Association Extension," filed on even date herewith, both of which are incorporated herein by reference.
An artificial vision system includes means (12, 14) for generating an image percept vector, means (16) for transforming this image percept vector into a feature vector, and means (16) for generating a response vector by multiplying the feature vector by a corresponding trained linkage matrix modeling a percept-response system.
A student neural network that is capable of receiving a series of tutoring inputs from one or more teacher networks to generate a student network output that is similar to the output of the one or more teacher networks. The tutoring inputs are repeatedly processed by the student until, using a suitable method such as back propagation of errors, the outputs of the student approximate the outputs of the teachers within a predefined range. Once the desired outputs are obtained, the weights of the student network are set. Using this weight set the student is now capable of solving all of the problems of the teacher networks without the need for adjustment of its internal weights. If the user desires to use the student to solve a different series of problems, the user only needs to retrain the student by supplying a different series of tutoring inputs.
A system for using a troubleshooting engine to assemble an interactive multimedia repair guide for assisting a service technician in the repair of a defective product, and a mehof of using the same. The system includes a dynamic knowledge database for storing product history records relating to defects reported concerning the product. The database also contains design information related to the product for correlation with the defect reports and use in analyzing future reported defects. A product performance counter (PPC) analysis module receives PPC data from an individual product, generates a PPC profile based on the received data, and compares the profile to stored design and historical PPC profiles to produce a weighted prediction report of likely defects. This report is transmitted to the troubleshooting engine, which directs a multimedia application to transmit diagnosis and repair instructions to a service center technician. The database is formed of separate portions, embodied at separate geographical locations and storing records in different languages.
