Iridocorneal Angle Image Stitching for Seamless Panoramic Views

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current medical imaging technologies fail to produce a single continuous panoramic image of the iridocorneal angle, resulting in discontinuous and difficult-to-use images when stitching multiple images side by side, which is undesirable for medical professionals.

Innovation Solution

A method involving image processing steps such as rotation, perspective correction, registration, positioning, and merging of adjacent images to create a seamless panoramic linear or annular image of the iridocorneal angle, utilizing techniques like non-reflective similarity transformations and color blending, with optional annotation and remapping.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If multiple images are placed side by side to create a panoramic view, then the coverage area of the iridocorneal angle is improved, but the continuity and usability of the image is worsened

Engineering Contradiction:
Improvecoverage areaVSAvoidimage continuity
Core Design Contradiction:
Area of stationary objectVSEase of operation

Solution Approach 1:

The patent merges multiple adjacent images of the iridocorneal angle through image processing techniques including registration, transformation, and blending. The method combines images taken at different angular positions into a single continuous panoramic representation, eliminating the discontinuities that would result from simply placing images side by side. This merging process maintains both the expanded coverage area and the visual continuity required for ease of operation.

Inventive Principle:
Principle #5Merging (Combining)

2Device complexity

If simple side-by-side placement of images is used, then the device complexity is reduced, but the image quality and usability for medical diagnosis is worsened

Engineering Contradiction:
Improveprocessing complexityVSAvoidimage quality
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The patent applies preliminary image processing actions to each image before final assembly, including registration to align anatomical features, geometric transformation to correct perspective distortions, and color normalization to ensure uniform appearance across the panoramic image. These preliminary actions are automatically performed by the image processing system, maintaining high image quality without requiring complex manual intervention.

Inventive Principle:
Principle #10Preliminary action

3Ease of operation

If panoramic stitching is implemented to provide continuous image, then the ease of operation is improved, but the processing time and computational resources are increased

Engineering Contradiction:
Improveimage usabilityVSAvoidprocessing time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The patent replaces manual image assembly operations with automated image processing algorithms. The system automatically performs registration, transformation, and blending of multiple images using computational methods, eliminating the need for time-consuming manual stitching while producing high-quality continuous panoramic images suitable for medical diagnosis.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP3785219B1Method of stitching a plurality of iridocorneal angle images
Publication Date: 2022.07.27 NIDEK CO LTD
  • EP3785219B1 patent drawingFigure 1
  • EP3785219B1 patent drawingFigure 2
  • EP3785219B1 patent drawingFigure 3

AI summary

The method serves for stitching a plurality of acquired images of an angular sector of an iridocorneal angle of an eye, and comprises the steps of: A) processing the acquired images, wherein each of the acquired image is rotated and/or perspectively corrected, B) registering the processed images, wherein for each couple of adjacent processed images a transformation is determined that correlates the couple, C) positioning the registered images, wherein for each registered image a corresponding relative positioning is carried out, and D) merging the repositioned images based on their relative positions; step B in particular is carried out taking into account the specificities of the images of the eye iridocorneal angle; in this way, a panoramic linear picture of the angular sector of iridocorneal angle of eye in the form of a rectangular picture is obtained; such panoramic linear picture may be processed through geometrical transformation and coordinate mapping so to obtain a panoramic circular picture of the whole iridocorneal angle of an eye in the form of an annular picture.