Microscopic Image Archive Tile Indexing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital microscopic images with high resolution and multiple focal planes require significant storage space and result in slow image processing, leading to cumbersome handling and navigation challenges.
Innovation Solution
An image file structure is generated with separate image and index bit streams, allowing direct access to image tiles at different resolution levels, utilizing an index bit stream for quick zooming and navigation through a two-stage accessing mechanism.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high resolution digital microscopic images with multiple focal planes are stored, then image quality and detail are improved, but storage space requirements and processing time increase significantly
Solution Approach 1:
The high-resolution microscopic image is divided into multiple smaller image tiles or blocks. Each tile can be independently stored, accessed, and processed. This segmentation allows the system to load only the specific tiles needed for current viewing rather than processing the entire large image, significantly reducing processing time while maintaining full image quality.
Solution Approach 2:
The patent introduces a hierarchical dimension to the image storage structure by creating multiple resolution levels (full resolution, reduced resolution, and thumbnail versions) of the same microscopic image. This dimensional approach allows users to navigate through different resolution layers, enabling quick overview at low resolution and detailed examination at full resolution without always loading the complete high-resolution data.
2Measurement precision
If complete high-resolution image data is loaded into memory, then image quality is maintained, but memory consumption and access latency increase
Solution Approach 1:
The patent pre-processes the microscopic images by generating multiple resolution levels and organizing them into a hierarchical structure before actual use. This preliminary action creates an optimized data structure that enables rapid access to different resolution levels without requiring real-time processing, thus reducing access latency while maintaining image quality when needed.
Solution Approach 2:
The image data is organized in a nested hierarchical structure where thumbnail images are contained within reduced-resolution images, which in turn are contained within the full-resolution image tiles. This nesting allows the system to progressively load data from lower to higher resolutions based on user needs, minimizing initial memory consumption while preserving the ability to access complete high-resolution data when required.
3Productivity
If multiple resolution levels are stored for each image, then navigation and zooming performance are improved, but data structure complexity increases
Solution Approach 1:
The complex multi-resolution data structure is segmented into standardized, uniformly sized image tiles at each resolution level. This tiling approach simplifies the overall structure by creating regular, predictable patterns that are easier to manage and access compared to storing complete images at multiple resolutions. The segmentation enables systematic organization and efficient retrieval.
Solution Approach 2:
The patent systematically varies the resolution parameter across different levels of the hierarchy while maintaining consistent tile dimensions and organizational patterns. By changing only the resolution parameter and keeping other structural elements uniform, the system achieves fast navigation through resolution levels without creating overly complex data structures, as the underlying tile-based organization remains constant.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for generating an image file archive (200), includes: providing an image (100) in a number m of resolution levels (110, 120, 130, 140), with each of the m resolution levels (110, 120, 130, 140) having a level-specific number ni of image tiles; providing an image bit stream (400), comprising the image in the m resolution levels (110, 120, 130, 140), with the image bit stream (400) having an image data entry (410-1, ..., 440-1) for each of the ni image tiles of each of the m resolution levels (110, 120, 130, 140); on the basis of the number m of resolution levels (110, 120, 130, 140) and the level-specific numbers ni of image tiles, providing an index bit stream (300), which comprises a succession of address data fields (310-1, ..., 340-1), with each address data field of the succession of address data fields (310-1, ..., 340-1) being indicative of a position of a particular image data entry of a particular image tile k at a particular resolution level i in the image bit stream; and combining the image bit stream (400) and the index bit stream (300) into the image file archive (200).