Depth Map Based Partial Image Blurring for UAV Cameras

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing techniques are unsuitable for cameras mounted on moving objects, such as unmanned aerial vehicles (UAVs), as they fail to effectively generate partially blurred images with a natural depth of field, especially for non-stereoscopic cameras.

Innovation Solution

The method involves generating a depth map of an image using pixel values representative of object distances, identifying different depths, and blurring pixels based on relative distances from a target depth, allowing for the creation of a partially blurred image even with 2D cameras on movable objects.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If manual pixel-blurring algorithm is applied to a single image, then processing can be done with simple camera systems, but the blurring appears unnatural and is not aesthetically appealing

Engineering Contradiction:
Improvecamera system complexityVSAvoidaesthetic quality of blurring
Core Design Contradiction:
Ease of manufactureVSDifficulty of detecting and measuring

Solution Approach 1:

The image is segmented into multiple depth layers using depth map generation. Each layer corresponds to objects at different distances from the camera, allowing selective blurring application based on depth information rather than uniform manual blurring across the entire image.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces depth as an additional dimension by generating a depth map that provides three-dimensional distance information for each pixel. This enables the system to apply blurring based on spatial depth relationships, transforming the two-dimensional image processing into a three-dimensional depth-aware process that produces natural-looking depth of field effects.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Difficulty of detecting and measuring

If multiple images from different perspectives are processed, then natural depth of field blur is achieved, but the technique is unsuitable for cameras that move significantly between captured images

Engineering Contradiction:
Improvenatural appearance of blurringVSAvoidcompatibility with moving camera systems
Core Design Contradiction:
Difficulty of detecting and measuringVSAdaptability or versatility

Solution Approach 1:

The patent implements a dynamic approach by tracking features across multiple frames and continuously updating the depth map as the camera moves. The system adapts to camera motion by re-identifying features in new positions and recalculating depth information, allowing the partial blurring effect to be maintained throughout video sequences or continuous imaging scenarios.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system uses feedback from feature tracking and depth map generation to continuously refine the blurring application. By monitoring feature positions across frames and updating depth information dynamically, the system adjusts the blurring parameters based on current scene geometry and camera position, ensuring natural-looking results even during significant camera movement.

Inventive Principle:
Principle #23Feedback

3Difficulty of detecting and measuring

If depth map generation and depth-based blurring is implemented, then natural depth of field is achieved, but processing complexity increases

Engineering Contradiction:
Improvequality of depth of field effectVSAvoidimage processing complexity
Core Design Contradiction:
Difficulty of detecting and measuringVSDevice complexity

Solution Approach 1:

The patent introduces a depth map as an intermediary data structure that stores depth information for each pixel. This intermediate representation serves as a bridge between the input images and the final blurred output, enabling the system to apply complex depth-based blurring operations efficiently by referencing pre-computed depth values rather than performing complex geometric calculations during the actual blurring process.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3488603B1Methods and systems for processing an image
Publication Date: 2021.07.28 SZ DJI TECH CO LTD
  • EP3488603B1 patent drawingFigure 1A
  • EP3488603B1 patent drawingFigure 1B
  • EP3488603B1 patent drawingFigure 1C

AI summary

Methods and systems for processing an image having a first set of pixels. The method may include generating a depth map of the image, the depth map including a second set of pixel values representative of distances of objects in the image. The method may further include identifying a plurality of different depths at which objects are located in the image based on the depth map, and using the depth map to determine a relative distance between one identified depth in the plurality of different depths and each of the other identified depths in the plurality of different depths. Finally, the method may include blurring pixels in the first set of pixels based on each determined relative distance.