Image Recognition Device Using Visible and Infrared Pixels

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional image recognition technologies face challenges in maintaining high recognition accuracy due to artifacts present in High Dynamic Range (HDR) images.

Innovation Solution

The proposed image recognition device employs an imaging unit that captures multiple images using visible light and infrared light imaging pixels at the same exposure timing, generating image data that is then processed by a recognition unit to recognize subjects without the influence of HDR artifacts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If HDR imaging is used to capture images with high dynamic range, then the imaging capability in varying light conditions is improved, but recognition accuracy decreases due to artifacts in HDR images

Engineering Contradiction:

Improveimaging capability in varying light conditionsVSAvoidrecognition accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent divides the imaging system into multiple independent imaging units, each capturing images at the same exposure timing. These segmented image data are then processed separately by the recognition unit, allowing the system to maintain HDR imaging capability while avoiding artifact generation that would reduce recognition accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent extracts and eliminates HDR artifacts from the image data by using multiple imaging units that capture images simultaneously at the same exposure timing. This extraction of artifact-free data allows the recognition unit to process clean image information, thereby improving recognition accuracy while preserving the adaptability of HDR imaging.

Inventive Principle:

Principle #2Taking out (Extraction)

2Adaptability or versatility

If multiple images are synthesized to generate HDR images, then the dynamic range is improved, but artifacts are introduced that harm recognition accuracy

Engineering Contradiction:

Improvedynamic rangeVSAvoidHDR artifacts

Core Design Contradiction:

Adaptability or versatilityVSObject-generated harmful factors

Solution Approach 1:

The patent converts the potential harm of artifact generation into a benefit by using multiple imaging units to capture images at the same exposure timing. Instead of synthesizing images that generate artifacts, the system uses parallel capture to obtain multiple clean image data sets, which are then processed to achieve high dynamic range without artifacts, thereby improving both adaptability and eliminating harmful factors.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enhances the recognition accuracy of subjects by eliminating the negative impact of HDR artifacts and allows for improved image processing across varying light conditions.

Implementation Method 1

imaging pixels which receive visible light and imaging pixels which receive infrared light

Methodology Applied

Scientific EffectVisible light detection: Light

Implementation Method 2

imaging pixels which receive infrared light

Methodology Applied

Scientific EffectInfrared light detection: Infrared Radiation

Data Source

PatentUS12302004B2Image recognition device and image recognition method

Publication Date: 2025.05.13 SONY SEMICON SOLUTIONS CORP

AI summary

Provided are an image recognition device and an image recognition method which can improve recognition accuracy of a subject. An image recognition device (an image sensor 1) according to the present disclosure includes an imaging unit (10) and a recognition unit (14). The imaging unit (10) uses the imaging pixels (R, Gr, Gb, B) which receive visible light and imaging pixels (IR) which receive infrared light, and images a plurality of images at the same exposure timing in one frame period to generate image data. The recognition unit (14) recognizes a subject from each of the image data.