On-Demand Image Tile Extraction from Compressed Hierarchical Layers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for downloading aerial images from servers are inefficient due to the need to transfer and decompress large compressed image files, resulting in significant delays before the images can be viewed at desired resolution and quality.

Innovation Solution

A method where a client requests and downloads only the specific image tile from a compressed image file at the desired resolution layer, allowing for on-demand decompression and transfer of the requested tile directly from the server, eliminating the need for pre-computed tiles and caching, and enabling flexible image organization and display.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If the whole compressed image file is downloaded from the server to the client and then decompressed, then the complete image is available for viewing, but large amounts of data must be transferred and processed, resulting in significant delays

Engineering Contradiction:
Improveamount of image data transferredVSAvoidtime to download and decompress image
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The patent divides the complete image into multiple image tiles organized in a hierarchical structure with different resolution layers. Instead of transferring the entire compressed image file, only the specific tile(s) at the required resolution layer are extracted and transferred to the client. This segmentation dramatically reduces the quantity of data transferred while maintaining the ability to display complete images by assembling multiple tiles.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts only the necessary image tile data from the compressed image file at the server side, based on the client's request for specific coordinates and resolution layer. This extraction eliminates the need to transfer unnecessary image data, reducing both the amount of data transferred and the decompression time required at the client side.

Inventive Principle:
Principle #2Taking out (Extraction)

2Productivity

If pre-computed image tiles are stored in a tile cache on the server, then subsequent tile requests can be fulfilled faster, but the server requires significant storage resources and pre-computation time

Engineering Contradiction:
Improvespeed of tile deliveryVSAvoidserver storage capacity for cached tiles
Core Design Contradiction:
ProductivityVSVolume of stationary object

Solution Approach 1:

The patent performs preliminary organization of image data into a hierarchical tile structure with multiple resolution layers during the image ingestion phase. However, unlike traditional caching approaches, it does not pre-compute and store decompressed tiles. Instead, the server maintains the compressed image file in an optimized hierarchical format, allowing rapid extraction of any tile on-demand without requiring extensive cache storage.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The server extracts and decompresses individual tiles on-demand from the hierarchical compressed image structure when requested by clients. This self-service approach eliminates the need for maintaining large pre-computed tile caches, as the server can generate any required tile dynamically from the source compressed image file using the hierarchical organization.

Inventive Principle:
Principle #25Self-service

3Speed

If streaming protocols with tile evolution are used to download low quality versions first, then initial display is faster, but there is still a delay before higher quality versions are available

Engineering Contradiction:
Improveinitial image display speedVSAvoiddelay before desired quality is achieved
Core Design Contradiction:
SpeedVSLoss of time

Solution Approach 1:

The patent implements a dynamic tile delivery system where the server can respond to client requests for different resolution layers of the same image tile. The hierarchical structure allows the server to provide lower resolution tiles quickly for initial display, then dynamically deliver higher resolution versions of the same tile coordinates when requested, eliminating the sequential delay inherent in traditional streaming protocols.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent adds a resolution layer dimension to the traditional single-resolution tile model. By organizing tiles in a hierarchical structure with multiple resolution layers (e.g., LOD0, LOD1, LOD2), the system can satisfy different quality requirements for the same spatial coordinates without requiring separate download sequences, enabling flexible quality switching for the same tile location.

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

Data Source

PatentEP3547165B1Image tile request service
Publication Date: 2023.06.07 ROSEMOUNT AEROSPACE LIMITED
  • EP3547165B1 patent drawingFigure 1
  • EP3547165B1 patent drawingFigure 2
  • EP3547165B1 patent drawingFigure 3

AI summary

A method of obtaining an image tile from a server by a client includes a client (7) submitting a request to a server for extracting an image tile (6) from a compressed image file (4). The server controls access to the compressed image file, the compressed image file contains multiple different resolution image layers (1, 2, ..., n) and the image tile is to be extracted from one of these image layers. The request includes the identity of the image layer (5) and the coordinates of the image tile in the image layer. The server extracts the requested image tile directly from the image layer in the compressed image file using the request and transfers image tile to the client at the decompressed image resolution and quality such that the image tile can be displayed on a display (8) of the client.