20230199280. MACHINE LEARNING DEVICE AND IMAGE PROCESSING DEVICE simplified abstract (JVCKENWOOD Corporation)

MACHINE LEARNING DEVICE AND IMAGE PROCESSING DEVICE

Organization Name

JVCKENWOOD Corporation

Inventor(s)

Hideki Takehara of Yokohama-shi (JP)

Shingo Kida of Yokohama-shi (JP)

Yincheng Yang of Yokohama-shi (JP)

MACHINE LEARNING DEVICE AND IMAGE PROCESSING DEVICE - A simplified explanation of the abstract

This abstract first appeared for US patent application 20230199280 titled 'MACHINE LEARNING DEVICE AND IMAGE PROCESSING DEVICE

Simplified Explanation

The abstract describes a system that uses machine learning to generate visible light images from far-infrared images. It also includes a trained identification model to determine if the far-infrared image was captured in a specific time zone.

• The system acquires a far-infrared image and a visible light image at different predetermined time zones.
• A generative adversarial network is used to train a visible light image generation model with the acquired images.
• The trained generation model can then generate visible light images from far-infrared images captured in a different time zone.
• The system also generates a trained identification model to identify if a far-infrared image was captured in the specific time zone.

Potential Applications

• Surveillance systems: Generating visible light images from far-infrared images can enhance surveillance capabilities in low-light conditions.
• Medical imaging: Converting far-infrared images to visible light images can aid in medical diagnosis and treatment.
• Environmental monitoring: The technology can be used to analyze far-infrared images captured at different times to study changes in temperature and other environmental factors.

Problems Solved

• Limited visibility in low-light conditions: The technology allows for the generation of visible light images from far-infrared images, improving visibility in dark environments.
• Time zone identification: The trained identification model can determine if a far-infrared image was captured in a specific time zone, providing valuable information for various applications.

Benefits

• Enhanced image visibility: The system enables the generation of visible light images from far-infrared images, improving the clarity and detail of captured images.
• Time zone identification: The trained identification model can accurately determine the time zone of a captured far-infrared image, aiding in data analysis and interpretation.

Original Abstract Submitted

a far-infrared image training data acquisition unit acquires a far-infrared image in a first predetermined time zone. a visible light image training data acquisition unit acquires a visible light image in a second predetermined time zone. a visible light image generation model training unit machine-learns the far-infrared image in the first predetermined time zone and the visible light image in the second predetermined time zone as training data by a generative adversarial network, and generates a trained generation model, which generates the visible light image in the second predetermined time zone from the far-infrared image in the first predetermined time zone. through machine learning by a generative adversarial network, the visible light image generation model training unit further generates a trained identification model, which identifies whether or not the far-infrared image is a far-infrared image captured in the first predetermined time zone.