Signal Processing Apparatus for AI-Enabled Imaging Systems

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing imaging systems without AI processing functions face challenges in implementing AI processing capabilities, which limits their functionality and efficiency in applications such as surveillance.

Innovation Solution

A signal processing apparatus that includes an AI processing part, an image processing part, a control part, and a transmission part, allowing for the addition of AI processing functions to imaging systems without such capabilities. This apparatus performs AI processing on input images using an AI model, selectively acquires different types of images, and transmits processing results externally.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If AI processing functions are added to imaging systems, then functionality and versatility are improved, but device complexity increases

Engineering Contradiction:
ImproveAI processing capabilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent introduces a signal processing apparatus as an intermediary component that connects to the imaging system via a standard interface. This mediator performs AI processing externally, allowing the imaging system to gain AI capabilities without internal modification. The signal processing apparatus receives captured images, applies AI models, and returns processed results, thus adding functionality while keeping the imaging system itself simple.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If AI processing is performed on captured images, then productivity is improved, but loss of information increases due to data transmission requirements

Engineering Contradiction:
Improveprocessing efficiencyVSAvoiddata communication overhead
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent extracts the AI processing function from the imaging system and places it in a separate signal processing apparatus. Only the necessary processed results are transmitted back to the imaging system, rather than transmitting all raw captured images for processing. This extraction approach minimizes data communication overhead while maintaining high processing efficiency.

Inventive Principle:
Principle #2Taking out (Extraction)

3Adaptability or versatility

If multiple AI models are supported, then adaptability is improved, but device complexity increases

Engineering Contradiction:
ImproveAI model selection flexibilityVSAvoidmodel management complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic AI model selection where the signal processing apparatus can load, switch, and manage multiple AI models based on processing requirements. The system dynamically adjusts which AI model is active without requiring permanent storage or complex internal architecture for each model. This dynamic approach enables multi-model support while keeping the device structure relatively simple.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS20250165746A1Signal processing apparatus, signal processing method, data structure, and data generation method
Publication Date: 2025.05.22 SONY SEMICON SOLUTIONS CORP
  • US20250165746A1 patent drawing
  • US20250165746A1 patent drawing
  • US20250165746A1 patent drawing

AI summary

A signal processing apparatus is provided which adds AI processing functions to an imaging system without such functions while improving the degree of freedom of AI processing that can be implemented by the imaging system.A signal processing apparatus according to the present technology includes an AI processing part configured to perform AI processing on an input image by using an AI model, an image processing part capable of selectively acquiring a different type of image as an image based on a captured image input from an outside, a control part configured to control settings of the image processing part on the basis of image designation data for designating an image to be obtained by the image processing part, and to control settings of the AI model for the AI processing part on the basis of AI model designation data for designating an AI model to be set to the AI processing part, and a transmission part configured to transmit information indicative of a result of the AI processing performed by the AI processing part to the outside.