Image Registration Boundary Display for Handheld Camera Alignment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for initial registration between a target object in a video and a simulation image are challenging due to difficulty in holding a camera still, making it hard to manually align the target object and simulation image, especially when the camera is handheld, leading to movement of the target object in the displayed video.

Innovation Solution

An image registration device that acquires a video and a simulation image, displays a boundary image instead of the simulation image during initial registration, allowing for easier alignment by enabling parallel movement, rotational adjustment, and enlargement/reduction of the boundary image to match the target object, facilitating first and second registrations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If a simulation image is displayed during initial registration, then registration information is provided, but the simulation image obstructs the target object making alignment difficult

Engineering Contradiction:
Improveregistration informationVSAvoidalignment ease
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The display is segmented into multiple regions: a first region displaying the video with the target object, and a second region displaying the simulation image with registration marks. This spatial segmentation allows the operator to view both the target object and registration information simultaneously without mutual obstruction, enabling easy alignment while maintaining access to registration data.

Inventive Principle:
Principle #1Segmentation

2Ease of operation

If the camera is held by hand for imaging, then mobility is improved, but the target object moves in the displayed video making registration difficult

Engineering Contradiction:
Improvecamera mobilityVSAvoidregistration precision
Core Design Contradiction:
Ease of operationVSManufacturing precision

Solution Approach 1:

The system provides real-time visual feedback by displaying both the video feed and simulation image with registration marks simultaneously on the display device. This allows the operator to see the current alignment status and make immediate adjustments to the camera position or orientation, enabling precise registration even while手持ing the camera for mobility.

Inventive Principle:
Principle #23Feedback

3Loss of information

If the simulation image is displayed overlapped on the target object, then spatial relationship is shown, but the target object cannot be seen clearly for alignment

Engineering Contradiction:
Improvespatial relationship informationVSAvoidtarget object visibility
Core Design Contradiction:
Loss of informationVSDifficulty of detecting and measuring

Solution Approach 1:

The display is divided into different spatial regions: the first region shows the video with the target object, and the second region shows the simulation image with registration marks. This dimensional separation in the display space allows both the target object and simulation image to be visible simultaneously without overlapping, enabling clear detection of the target object while maintaining spatial relationship information.

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

Data Source

PatentUS10631948B2Image alignment device, method, and program
Publication Date: 2020.04.28 FUJIFILM CORP
  • US10631948B2 patent drawing
  • US10631948B2 patent drawing
  • US10631948B2 patent drawing

AI summary

There is provided an image registration device, method, and program that enable easy initial registration between a target object included in a video and a simulation image. A first registration unit performs first registration that is initial registration between an intraoperative video and a simulation image. At this time, a boundary image showing the boundary of the simulation image is displayed on a display so as to be superimposed on the intraoperative video. An operator performs registration between a target object included in the intraoperative video and the boundary image. After the end of the first registration, a second registration unit performs second registration between the simulation image and the target object included in the intraoperative video based on the result of the first registration.