Granted Patent B2

US 12,599,305 · App. 19/031,061 · Granted Apr 14, 2026

Inventor: Mohammed N. Islam (Ann Arbor, MI)

Assignee: OMNI MEDSCI, INC.

A61B5/0088A61B5/0013A61B5/0022A61B5/0075A61B5/0086A61B5/14532A61B5/14546A61B5/1455A61B5/4547A61B5/6801A61B5/7203A61B5/7257A61B5/7405A61B5/742A61C19/04G01J3/02G01J3/0218G01J3/108G01J3/14G01J3/28G01J3/2823G01J3/42G01J3/453G01N21/35G01N21/3504G01N21/3563G01N21/359G01N21/39G01N21/88G01N33/02G01N33/025G01N33/15G01N33/442G01N33/49G16H40/67G16Z99/00G01N2201/129H01S3/302

Abstract

Claims (20)

1 . A remote sensing system comprising:

an array of laser diodes configured to generate a light having an initial light intensity and one or more optical wavelengths, wherein the array of laser diodes comprises a plurality of emitters;

wherein at least a portion of the one or more optical wavelengths is a near-infrared wavelength between 700 nanometers and 2500 nanometers, wherein at least a portion of the array of laser diodes comprises one or more Bragg reflectors, wherein the at least a portion of the array of laser diodes is further configured to be modulated with a pulsed output with a pulse duration of approximately 0.5 to 2 nanoseconds and a pulse repetition rate between about 10 Megahertz and 1 Gigahertz, wherein the array of laser diodes is further coupled to driver electronics;

a first lens configured to receive at least a portion of the light from the array of laser diodes and configured to direct at least some of the portion of the light from the array of laser diodes to an object;

a photodiode array, at least one second lens and one or more spectral filters in front of at least a part of the photodiode array, wherein the photodiode array is further coupled to a processor, and wherein the photodiode array comprises a plurality of pixels coupled to CMOS transistors;

wherein the photodiode array is configured to receive at least a portion of light reflected from the object, and wherein the photodiode array is further configured to be synchronized to the at least a portion of the array of laser diodes comprising one or more Bragg reflectors;

wherein the processor is configured to perform a time-of-flight measurement based on a time difference between a first time in which the at least a portion of the array of laser diodes generate light and a second time in which the photodiode array receives a received portion of light reflected from the object;

wherein the processor is further configured to perform the time-of-flight measurement at least in part by measuring a temporal distribution of photons in the received portion of light;

wherein the processor is configured to be coupled to a non-transitory computer readable medium;

wherein at least a portion of the time-of-flight measurement is configured to serve as at least one input to a multi-modal generative artificial intelligence model configured to communicate with or operate at least in part on the processor; and

wherein the multi-modal generative artificial intelligence model comprises a vision transformer configured to analyze an item in an input video or an input image, and wherein the vision transformer comprises self-attention and positional encoding layers.

2 . The remote sensing system of claim 1 , wherein the remote sensing system including the processor is configured to be coupled to a camera configured to be coupled to a third lens and the processor, the camera configured to capture one or more images including at least a part of the object;

wherein the processor is configured to combine at least a portion of the one or more images and the at least a portion of the time-of-flight measurement to create a combined portion; and

wherein the combined portion is configured to serve as at least another input to the multi-modal generative artificial intelligence model.

3 . The remote sensing system of claim 1 , wherein the multi-modal generative artificial intelligence model is configured to be trained or fine-tuned using reinforcement learning, reinforcement fine-tuning, prompt tuning, or supervised fine-tuned using a specific set of labeled datasets.

4 . The remote sensing system of claim 1 , wherein the multi-modal generative artificial intelligence model is configured to perform data fusion by combining the at least a portion of the time-of-flight measurement with data from other sensors.

5 . The remote sensing system of claim 1 , wherein the vision transformer is configured to be trained on datasets comprising one or more training images or one or more training videos, and wherein a prompt to the multi-modal generative artificial intelligence model comprises the input image or the input video; and

wherein the multi-modal generative artificial intelligence model is further configured to perform anomalous occurrence detection, wherein the one or more training images or the one or more training videos correspond to normal occurrences that are configured to fine-tune the multi-modal generative artificial intelligence model, and then the anomalous occurrence detection is configured to detect differences in the input image or input video from normal occurrences that fall outside a threshold value.

6 . The remote sensing system of claim 1 , wherein the multi-modal generative artificial intelligence model further comprises natural language processing configured to accept and analyze voice prompts or voice-based interactions.

7 . The remote sensing system of claim 1 , wherein the remote sensing system including the processor is configured to be further coupled to an agent, a robot or an industrial automated equipment or vehicle, wherein an output from the multi-modal generative artificial intelligence model is configured to serve as a guidance input to the agent, the robot or the industrial automated equipment or vehicle.

Assignments (1)

Continuity (37)

References Cited (400)