Parameter-Free Area of Interest Discovery via Density Estimation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for discovering and recommending areas of interest (AOI) in smaller cities face challenges due to lack of objective information sources and subjective recommendations, with existing clustering techniques like K-means, Mean Shift, and DBSCAN requiring parameter tuning and resulting in varying region sizes and low photo density issues.

Innovation Solution

A method and system that processes user requests for recommended AOI by computing region representations based on geo-location coordinates, using unique photograph density calculations, kernel density estimations, and watershed segmentation to identify optimal areas of interest, ranking them by density parameters, and filtering data based on user-defined constraints like time and transportation, without requiring parameter tuning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If DBSCAN clustering technique is used for AOI discovery, then regions can be identified based on photo density, but the discovered regions greatly vary in size depending on photo density and require parameter tuning

Engineering Contradiction:

Improvephoto density measurementVSAvoidparameter tuning complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs self-service by automatically determining optimal clustering parameters through evaluation of multiple candidate parameter sets against the photo data, eliminating the need for manual parameter tuning by domain experts while maintaining high measurement precision for photo density

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system applies parameter changes by testing multiple different parameter configurations (radius values, minimum point thresholds) and selecting the optimal set that best represents the photo density distribution, thereby resolving the contradiction between precision and complexity

Inventive Principle:

Principle #35Parameter changes

2Productivity

If DBSCAN clustering is applied with fixed parameters, then clustering can be performed efficiently, but regions of interest may be missed in low density areas

Engineering Contradiction:

Improveclustering processing speedVSAvoidregion discovery completeness

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system applies dynamics by making the clustering parameters adaptive rather than fixed - it dynamically adjusts radius and minimum point thresholds based on local photo density characteristics, ensuring both efficient processing and complete region discovery across varying density areas

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs preliminary action by pre-evaluating multiple parameter configurations and selecting the optimal set before actual clustering execution, which ensures both reliability in region discovery and efficiency in the main processing phase

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If K-means clustering is used for POI discovery adapted to AOI, then the whole set of photos is partitioned into subsets, but every single point is assigned to one cluster resulting in large cells with low photo density

Engineering Contradiction:

Improveclustering operation simplicityVSAvoidregion boundary precision

Core Design Contradiction:

Ease of operationVSManufacturing precision

Solution Approach 1:

The system applies inversion by reversing the traditional K-means approach - instead of assigning every point to exactly one cluster, it allows points to belong to multiple overlapping clusters or no cluster at all, thereby creating more precise region boundaries that better reflect actual photo density patterns

Inventive Principle:

Principle #13The other way round (Inversion)

4Measurement precision

If Mean Shift technique is used for POI discovery, then areas with high photo density are identified, but the method does not allow identifying an AOI around or in proximity to the local peak

Engineering Contradiction:

Improvephoto density peak detectionVSAvoidAOI coverage area

Core Design Contradiction:

Measurement precisionVSArea of stationary object

Solution Approach 1:

The system applies merging by combining multiple clustering results from different parameter configurations, thereby merging the benefits of precise peak detection with expanded area coverage to identify complete AOI regions around and beyond local density peaks

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS9971775B2Method of and system for parameter-free discovery and recommendation of areas-of-interest

Publication Date: 2018.05.15 Y E HUB ARMENIA LLC

AI summary

A method and system for processing a user request for a recommended area of interest includes the steps of receiving the request including an indication of an electronic device geo-location and a user defined search constraint; receiving data associated with photographs associated with geo-objects, the data comprising geo-location coordinates of the photographs, the geo-location coordinates of the photographs being in proximity with the device geo-location; computing a plurality of region representations based on the geo-location coordinates of the photographs, each region representation being associated with a unique photograph density calculation parameter, the computing comprises determining a potential area of interest in each region representation, each region representation being a candidate for an optimal region representation; determining the optimal region representation based on the user defined search constraint; and displaying to the user the recommended area of interest that corresponds to the potential area of interest of the optimal region representation.