Gaming Environment for Real-World Object Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing and analysis technologies face challenges in reliably and scalably identifying objects in dense urban settings from diverse datasets collected by various sensors, leading to inefficiencies in creating accurate location information databases.
Innovation Solution
A system and method that utilizes an electronic gaming environment to simulate real-world locations, where players interact with and confirm the presence of objects, using game features and machine learning to generate updated maps by tracking player actions and exporting data to a database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional image processing and analysis technologies are used to identify objects in dense urban settings, then automated object detection is attempted, but the reliability and accuracy of identification deteriorates due to densely connected objects appearing indistinguishable from one another
Solution Approach 1:
The patent introduces human players as an intermediary between automated detection systems and final object identification. Players view aerial imagery in a gaming environment and manually identify/confirm objects, bridging the gap between automated detection capabilities and reliable identification in dense urban settings where machines struggle to distinguish closely connected structures
Solution Approach 2:
The system creates a virtual copy of real-world aerial imagery within a gaming environment. This copied visual data allows players to interact with and identify objects without affecting real-world operations, enabling scalable human verification while maintaining the original data integrity
2Reliability
If more analysts are hired to interpret sensor data and identify objects manually, then identification accuracy improves, but the cost and time required to process data increases significantly
Solution Approach 1:
The patent makes human analysts universally applicable by deploying them through a gaming platform that can simultaneously engage many players worldwide. Instead of hiring individual analysts for specific projects, the system creates a scalable network of voluntary players who all contribute to the same identification task, dramatically increasing processing capacity without proportional cost increases
Solution Approach 2:
The system leverages players' voluntary participation and intrinsic motivation to perform data interpretation tasks. Players self-select into the gaming environment and voluntarily contribute their identification skills without requiring direct compensation, eliminating the need for organizational overhead associated with hiring and managing professional analysts
3Reliability
If human analysts are used to identify objects in sensor data, then identification accuracy improves, but the scalability and economic viability of the solution deteriorates due to limited availability of analysts
Solution Approach 1:
The patent transitions human analysis from a linear, sequential process (one analyst at a time) to a parallel, multi-dimensional process by deploying numerous players simultaneously through a digital gaming platform. This dimensional shift from physical to virtual space allows unlimited scaling without being constrained by geographical or organizational boundaries
4Area of stationary object
If conventional automated systems are used for wide area surveillance, then coverage area increases, but the ability to distinguish one structure from another deteriorates when objects appear densely connected
Solution Approach 1:
The patent segments the surveillance task into two distinct phases: automated broad-area coverage for initial detection, and human-focused verification for precise structure distinction. The gaming platform divides the large-scale imagery into manageable sections that players can systematically review, allowing both wide coverage and high precision to coexist
Data Source
AI summary
A system and processes provide a solution to object identification in images using a game environment to confirm the presence (or absence) of suspected points of interest in the real-world. A game environment uses data gathered by real-world sensors to replicate a real-world location as a gaming environment. Unconfirmed suspected points of interest (locations and/or objects) may be placed into the game environment. Various types of game play may be used to have players interact with the game environment. The actions by players may confirm whether a suspected point of interest is actually present in the real-world. Confirmation data of the presence of points of interest may be forwarded to an external database which may update a map file based on the data provided by player interaction in the game.