Interaction Index Map Generation for Image Retrieval

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for obtaining information from images are inconvenient, as they require manual labeling or complex image identification processes, which are time-consuming and inefficient, especially when dealing with panoramic images or detailed content retrieval.

Innovation Solution

An electronic calculating apparatus and method that automatically generates an interaction index map of an image by matching features with reference images, transforming candidate reference images, and creating an interaction index map to facilitate user interactions through a displaying device using side-information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual labeling is used to obtain image information, then information can be obtained, but the process is time-consuming and inefficient

Engineering Contradiction:

Improveefficiency of obtaining image informationVSAvoidtime required for manual labeling

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system enables automatic self-labeling by using the image itself as reference. The processor automatically generates interaction index maps by matching features between the input image and reference images stored in the database, eliminating the need for manual labeling while maintaining information accuracy

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system creates a copy of the image (input image) and uses it as reference material. By copying the image content into the database as a reference, the system can then automatically match and generate interaction index maps without requiring manual intervention or additional reference materials

Inventive Principle:

Principle #26Copying

2Measurement precision

If Augmented Reality image identifying technology is used, then image contents can be identified, but computational complexity and power consumption increase

Engineering Contradiction:

Improveimage content identification accuracyVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system extracts only the essential feature information from the full image data. By extracting key features and using them for matching against reference images, the system achieves accurate content identification while reducing computational complexity compared to processing the entire image through complex AR algorithms

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs preliminary action by pre-storing reference images and their interaction index maps in the database before actual processing. This pre-processing allows the system to quickly match features during operation without performing complex real-time analysis, thereby reducing computational complexity while maintaining identification accuracy

Inventive Principle:

Principle #10Preliminary action

3Adaptability or versatility

If Augmented Reality image identifying technology is used, then image contents can be identified, but it can only handle near objects and not panoramic images

Engineering Contradiction:

Improvecapability to handle different image typesVSAvoidlimitation in image processing scope

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The system achieves universality by using a database of reference images that can accommodate various image types including panoramic images. The feature matching mechanism is designed to work with different image characteristics, allowing the same system to process both near objects and panoramic images without requiring separate specialized processing

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Quantity of substance

If searching engine is used to retrieve image information, then overall information can be obtained, but detailed content information is not available

Engineering Contradiction:

Improveamount of image informationVSAvoiddetail level of image content

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The system segments the image information into detailed interaction index maps that correspond to specific regions and features. By dividing the overall image information into segmented interaction indices (such as object identification, location, characteristics), the system provides both comprehensive coverage and detailed content information simultaneously

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS9524303B2Electronic calculating apparatus for generating an interaction index map of an image, method thereof and non-transitory machine-readable medium thereof

Publication Date: 2016.12.20 INSTITUTE FOR INFORMATION INDUSTRY

AI summary

An electronic calculating apparatus for generating an interaction index map of an image, a method thereof and a non-transitory machine-readable medium thereof are provided. The electronic calculating apparatus includes a database and a processor electrically coupled to the database. The database stores a plurality of reference images. The processor sets at least one feature of the image, selects at least one candidate reference image from the plurality of reference images according to the at least one feature by a matching calculating procedure, transforms the at least one candidate reference image into at least one transformed candidate reference image, and generates a specific interaction index map of the image according to at least one interaction index map of the at least one transformed candidate reference image so that a displaying device executes the corresponding operation according to a user instruction by the specific interaction index map and side-information.