Variable focus liquid lens optical assembly for value chain networks

Patent metadata

Field	Value
Publication number	US-12450581-B2
Application number	US-202217683198-A
Country	US
Kind code	B2
Filing date	Feb 28, 2022
Priority date	Dec 18, 2020
Publication date	Oct 21, 2025
Grant date	Oct 21, 2025

Abstract

A dynamic vision system includes a variable focus liquid lens optical assembly. The dynamic vision system includes a control system configured to adjust one or more optical parameters and data collected from the variable focus liquid lens optical assembly in real time. The dynamic vision system includes a processing system that dynamically learns on a training set of outcomes, parameters, and data collected from the variable focus liquid lens optical assembly to train one or more machine learning models to recognize an object.

First claim

The invention claimed is:

1. A dynamic vision system comprising:
a variable focus liquid lens optical assembly;
a control system configured to adjust one or more optical parameters of the variable focus liquid lens optical assembly and data collected from the variable focus liquid lens optical assembly in real time; and
a processing system that dynamically learns on a training set comprising: (a) at least one of outcomes or optical parameters, and (b) data collected from the variable focus liquid lens optical assembly to train one or more machine learning models to recognize an object, wherein the one or more machine learning models are embodied on a semiconductor chip that is integrated into a device or a system that houses the variable focus liquid lens optical assembly.

2. The dynamic vision system of claim 1 wherein the variable focus liquid lens optical assembly is continuously adjusted by the control system based on at least one of environment factors or on feedback from the processing system to generate an object concept.

3. The dynamic vision system of claim 2 wherein the object concept includes contextual intelligence about at least one of the object or its environment and provides superior object recognition by the dynamic vision system.

4. The dynamic vision system of claim 1 wherein a first machine learning model is used to optimize collection of signals by the variable focus liquid lens optical assembly and a second machine learning model is used to operate on the signals to achieve a desired vision outcome.

5. The dynamic vision system of claim 1 wherein the processing system receives real-time adjustable data streams from the variable focus liquid lens optical assembly to at least one of: generate situational awareness or create out-of-focus images of the object so as to capture at least one of rich metadata or contextual intelligence about at least one of the object or its environment.

6. The dynamic vision system of claim 1 wherein the control system and the processing system are integrated with the variable focus liquid lens optical assembly.

7. The dynamic vision system of claim 1 wherein the optical parameters adjusted by the control system include at least one of: focal length, specularity, color, environment, or lens shape.

8. The dynamic vision system of claim 7 wherein the adjustment of the optical parameters results in a change to at least one of: spherical aberration, field curvature, coma, chromatic aberration, distortion, vignetting, ghosting, flaring, or diffraction.

9. The dynamic vision system of claim 1 wherein:
the processing system is configured to train the one or more machine learning models to derive a configuration of the variable focus liquid lens optical assembly; and
the configuration includes at least one of: liquid lens material, geometry, shape, optical properties, performance, or design.

10. The dynamic vision system of claim 1 wherein:
the one or more machine learning models are pre-trained on a separate system, such as in a cloud computing environment, such as using at least one of: a large training data set of visual information or outcomes, to perform a set of machine vision tasks, and
the one or more pre-trained machine learning models are deployed on a device or a system that includes the variable focus liquid lens optical assembly.

11. A dynamic vision system comprising:
an image processing system configured to collect data from a variable focus liquid lens optical assembly; and
an artificial intelligence system configured to learn on a training set of at least one of: outcomes, parameters, or data collected from the variable focus liquid lens optical assembly to recognize an object and to control the variable focus liquid lens optical assembly to optimize the collection of data for processing by the artificial intelligence system.

12. A vision system for dynamically learning an object concept about an object of interest, the vision system comprising:
a variable focus liquid lens assembly;
a control system configured to adjust one or more optical parameters of the variable focus liquid lens assembly in real time;
one or more vision sensors configured to capture a real-time pixel array based on data received from the variable focus liquid lens assembly in response to adjustments by the control system, wherein the real-time pixel array represents the object concept;
an adaptive intelligence system configured to process the object concept to build a three-dimensional representation of the object, wherein the adaptive intelligence system includes: a machine learning system configured to input the object concept into one or more machine learning models, and wherein the object concept is used as training data for the one or more machine learning models; and
an artificial intelligence system configured to make at least one of: classifications, predictions, or
one or more decisions relating to the object, including determining at least one of: position, orientation, or motion of the object.

13. The vision system of claim 12 , wherein the object concept includes contextual data about one or more of: the object or an environment of the object.

14. The vision system of claim 12 , wherein the adaptive intelligence system is configured to build the three-dimensional representation of the object directly from the object concept without generating an intermediate image.

15. A dynamic vision system comprising:
a variable focus liquid lens optical assembly;
a control system configured to adjust one or more optical parameters of the variable focus liquid lens optical assembly and data collected from the variable focus liquid lens optical assembly in real time; and
a processing system that dynamically learns on a training set comprising (a) at least one of outcomes or optical parameters, and (b) data collected from the variable focus liquid lens optical assembly to train one or more machine learning models to recognize an object, wherein a first machine learning model of the one or more machine learning models is used to optimize collection of signals by the variable focus liquid lens optical assembly and a second machine learning model of the one or more machine learning models is used to operate on the signals to achieve a desired vision outcome.

16. The dynamic vision system of claim 15 , wherein the desired vision outcome comprises at least one of: a recognition outcome, a classification outcome, or a prediction outcome.

17. The dynamic vision system of claim 15 , wherein the first machine learning model or the second machine learning model comprises a convolutional neural network (CNN).

18. A dynamic vision system comprising:
a variable focus liquid lens optical assembly;
a control system configured to adjust one or more optical parameters of the variable focus liquid lens optical assembly and data collected from the variable focus liquid lens optical assembly in real time; and
a processing system that dynamically learns on a training set comprising (a) at least one of outcomes or optical parameters, and (b) data collected from the variable focus liquid lens optical assembly to train one or more machine learning models to recognize an object, wherein the processing system receives real-time adjustable data streams from the variable focus liquid lens optical assembly to at least one of: generate situational awareness or create out-of-focus images of the object so as to capture at least one of rich metadata or contextual intelligence about at least one of the object or its environment.

Assignees

• Strong Force Vcn Portfolio 2019 Llc

Inventors

• Marinkovich Sava
• Cella Charles H
• Bliven Brent
• Dobrowitsky Joshua
• Cardno Andrew
• Sharma Kunal
• Kell Brad

Classifications

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N5/025

  Extracting rules from data · CPC title

• G06N3/088

  Non-supervised learning, e.g. competitive learning · CPC title

• G06N3/084

  Backpropagation, e.g. using gradient descent · CPC title