Dual-Camera Depth Calculation with Synthetic Noise Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional dual-camera systems for image processing in mobile terminals face challenges in achieving accurate depth information and good imaging effects, especially in low-light environments, leading to poor blurring effects.

Innovation Solution

The system employs a method where a high-resolution camera is used as the main camera in bright conditions and a high-ISO camera in low-light conditions, with both cameras working together to acquire multiple frames, perform synthetic noise reduction, and calculate depth information using the triangulation ranging principle for improved blurring processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a dual-camera system is used to acquire two photos for background blurring processing, then depth information can be calculated and blurring effect can be achieved, but in dark environments the imaging effect becomes poor and depth information accuracy decreases

Engineering Contradiction:
Improvedepth information accuracyVSAvoiddark environment imaging quality
Core Design Contradiction:
Measurement precisionVSIllumination intensity

Solution Approach 1:

The patent segments the imaging task by using different cameras for different purposes: one camera (first camera) is dedicated to capturing the main image, while another camera (second camera) is dedicated to capturing depth information. This segmentation allows each camera to be optimized for its specific function, and crucially, allows the main image camera to use multiple frames for noise reduction without being constrained by the depth sensing requirement.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary noise reduction on multiple frames captured by the first camera before selecting a basic frame. By pre-processing multiple frames to reduce noise, the system ensures that the selected basic frame has high quality even in dark environments, before it is used for subsequent depth calculation and blurring processing.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If multiple frames are captured and processed for noise reduction, then imaging effect improves, but processing time and complexity increase

Engineering Contradiction:
Improveimaging effect qualityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent performs noise reduction on multiple frames captured by the first camera before selecting a basic frame. By pre-processing multiple frames to reduce noise, the system ensures that the selected basic frame has high quality even in dark environments, before it is used for subsequent depth calculation and blurring processing.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent extracts only the necessary number of frames (n frames) from the multiple captured frames for noise reduction processing. Instead of processing all captured frames, the system selectively uses a subset of frames, which reduces processing time and computational complexity while still achieving effective noise reduction and improved imaging quality.

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If conventional dual-camera systems are used in dark environments, then depth information can be obtained, but the accuracy of depth information and image quality deteriorate

Engineering Contradiction:
Improvedepth information accuracyVSAvoiddark environment performance
Core Design Contradiction:
Measurement precisionVSIllumination intensity

Solution Approach 1:

The patent segments the imaging task by using different cameras for different purposes: one camera (first camera) is dedicated to capturing the main image, while another camera (second camera) is dedicated to capturing depth information. This segmentation allows each camera to be optimized for its specific function, and crucially, allows the main image camera to use multiple frames for noise reduction without being constrained by the depth sensing requirement.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the operating parameters of the camera system by using different ISO values for the two cameras. The first camera uses a first ISO value optimized for image quality, while the second camera uses a second ISO value optimized for depth sensing. This parameter differentiation allows each camera to perform its specific function optimally, especially in dark environments where ISO settings critically affect performance.

Inventive Principle:
Principle #35Parameter changes

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 imaging effect and depth information accuracy, overcoming the limitations of poor image quality and low focusing speed in conventional dual-camera systems, especially in dark environments.

Implementation Method 1

calculate depth information of the main image according to the main image and the auxiliary image

Methodology Applied
Scientific EffectTriangulation ranging principle: Parallax

Data Source

PatentEP3499863B1Method and device for image processing
Publication Date: 2021.12.22 SHENZHEN HEYTAP TECHNOLOGY CO LTD
  • EP3499863B1 patent drawingFigure 1~2
  • EP3499863B1 patent drawingFigure 3
  • EP3499863B1 patent drawingFigure 4

AI summary

A method for image processing includes the following actions. After n frames of first images shot by a main camera are acquired and an auxiliary image shot by an auxiliary camera is acquired, synthetic noise reduction is performed on the n frames of first images to obtain a frame of main image. Depth information of the main image is further calculated according to the main image and the auxiliary image. Therefore, blurring processing is performed on the main image according to the depth information of the main image to obtain a required second image.