Ensemble learning based image classification systems

• US 10,387,772 B1
• Filed: 10/22/2018
• Issued: 08/20/2019
• Est. Priority Date: 10/22/2018
• Status: Active Grant

• Current Assignee: Gyrfalcon Technology Inc.
• Original Assignee: Gyrfalcon Technology Inc.
• Inventors: Yang, Lin, Dong, Patrick Z., Young, Charles Jin, Dong, Jason Z., Lin, Michael, Sun, Baohua
• Primary Examiner(s): Dang, Duy M
• Application Number:
  US16/167,523
• Time in Patent Office: 302 Days
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06F 18/2411 based on the proximity to a decision surface, e.g. support vector machines
• CPC Class Codes: G06F 18/2413 based on distances to training or reference patterns
• CPC Class Codes: G06F 18/254 of classification results, e.g. of results related to same input data
• CPC Class Codes: G06F 18/259 Fusion by voting
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/20 Ensemble learning
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 3/063 using electronic means
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: G06N 7/00 Computing arrangements based on specific mathematical models
• CPC Class Codes: G06V 10/764 using classification, e.g. of video objects
• CPC Class Codes: G06V 10/809 of classification results, e.g. where the classifiers operate on the same input data
• CPC Class Codes: G06V 10/955 using specific electronic processors
• CPC Class Codes: G06V 20/00 Scenes; Scene-specific elements control of digital cameras H04N23/60

Abstract

An ensemble learning based image classification system contains multiple cellular neural networks (CNN) based integrated circuits (ICs) operatively coupling together as a set of base learners of an ensemble for an image classification task. Each CNN based IC is configured with at least one distinct deep learning model in form of filter coefficients. The ensemble learning based image classification system further contains a controller configured as a meta learner of the ensemble and a memory based data buffer for holding various data used in the ensemble by the controller and the CNN based ICs. Various data may include input imagery data to be classified. Various data may also include extracted feature vectors or image classification outputs out of the set of base learners. The extracted feature vectors or image classification outputs are then used by the meta learner to further perform the image classification task.

Claims

1. An ensemble learning based image classification system comprising:

a plurality of cellular neural networks (CNN) based integrated circuits (ICs) operatively coupling together as a set of base learners of an ensemble for an image classification task, each CNN based IC being configured with at least one of a plurality of distinct deep learning models;

a controller configured as a meta learner of the ensemble; and

a memory based data buffer for holding various data used in the ensemble by the controller and the plurality of CNN based ICs.

2. The ensemble learning based image classification system of claim 1, wherein the plurality of CNN based ICs is coupled together via a printed circuit board.

3. The ensemble learning based image classification system of claim 1, wherein the plurality of CNN based ICs is coupled together via a hub of dongles.

4. The ensemble learning based image classification system of claim 1, wherein all of the distinct deep learning models being trained using a common labeled database for a set of categories used in the image classification task.

5. The ensemble learning based image classification system of claim 1, wherein the various data comprises input imagery data to be classified.

6. The ensemble learning based image classification system of claim 5, wherein said each CNN based IC further comprising:

a plurality of cellular neural networks (CNN) processing engines operatively coupled to at least one input/output data bus, the plurality of CNN processing engines being connected in a loop with a clock-skew circuit, each CNN processing engine comprising;

a CNN processing block configured for simultaneously performing convolutional operations using the input imagery data and pre-trained filter coefficients of the at least one of the plurality of distinct deep learning models;

a first set of memory buffers operatively coupling to the CNN processing block for storing relevant portion of the input imagery data; and

a second set of memory buffers operative coupling to the CNN processing block for storing the pre-trained filter coefficients.

7. The ensemble learning based image classification system of claim 1, wherein the various data comprises extracted feature vectors out of the set of base learners.

8. The ensemble learning based image classification system of claim 7, wherein the extracted feature vectors are used by the meta learner to further perform the image classification task.

9. The ensemble learning based image classification system of claim 8, wherein the meta learner is configured with a machine learning model selected from the group consisting of (i) weighted voting, (ii) support vector machine, (iii) logistic regression, (iv) random forests, and (v) gradient boosting decision trees.

10. The ensemble learning based image classification system of claim 1, wherein the various data comprises image classification outputs out of the set of base learners.

11. The ensemble learning based image classification system of claim 10, wherein the image classification outputs are used by the meta learner to further perform the image classification task.

12. The ensemble learning based image classification system of claim 8, wherein the meta learner is configured with a machine learning model selected from the group consisting of (i) weighted voting, (ii) support vector machine, (iii) logistic regression, (iv) random forests, and (v) gradient boosting decision trees.

13. The ensemble learning based image classification system of claim 1, wherein the controller is further configured for performing computations of fully-connected layers.