ATM Motion Monitoring Using Skeleton Detection for Fraud Response
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional monitoring systems for money handling transaction apparatuses are inadequate in detecting and counteracting malicious activities, providing real-time operation support, and distinguishing between operators and bystanders, leading to insufficient security and usability.
Innovation Solution
A monitoring apparatus and method that utilize an imaging unit to detect joint positions, determine motions by comparing with predefined operator and bystander patterns, and generate instructions for response, including warnings, guidance, and security measures based on these determinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional image monitoring is used, then simple image recording is achieved, but malicious activity detection and real-time response capability are insufficient
Solution Approach 1:
The monitoring system is segmented into distinct functional modules: imaging unit for capture, posture detection unit for skeleton analysis, determination unit for motion classification, and instruction unit for response generation. Each module handles a specific aspect of monitoring, enabling comprehensive security analysis while maintaining modular architecture that manages complexity.
Solution Approach 2:
The determination unit acts as an intermediary between the posture detection unit and the instruction unit. It processes motion information by comparing detected postures against predefined operator motion patterns, classifying actions as normal or abnormal, and generating appropriate instructions. This intermediary layer enables intelligent decision-making without directly increasing system complexity.
2Reliability
If real-time monitoring and analysis is implemented, then malicious activity detection improves, but processing time and computational resources increase
Solution Approach 1:
Operator motion patterns and corresponding instructions are predefined and stored in advance. The determination unit compares real-time detected postures against these pre-established patterns, enabling rapid classification of actions without complex real-time analysis. This preliminary preparation significantly reduces processing time while maintaining high detection accuracy.
Solution Approach 2:
The system replaces complex real-time computational analysis with pattern matching against predefined motion templates. Instead of performing complex machine learning inference or deep analysis during real-time operation, the system uses straightforward comparison of detected skeleton positions against stored operator behavior patterns, substantially reducing computational burden and processing time.
3Device complexity
If operators and bystanders are treated indistinguishably, then monitoring coverage is simplified, but security against malicious activities decreases
Solution Approach 1:
The system applies different monitoring quality levels to different subjects. Operator motion patterns are captured and stored for reference, while bystander actions are monitored against this baseline. The determination unit specifically analyzes motions of operators using stored pattern data, while treating bystander behaviors differently. This localized differentiation enables effective security monitoring without uniformly complex analysis of all detected persons.
Data Source
AI summary
An ATM that internally has a function as a monitoring apparatus includes: an imaging unit that includes, in an imaging range, an operator of a money handling transaction apparatus and surroundings of the operator; an posture detection unit that detects a joint position related to a skeleton in an image of a person included in an imaging result of the imaging unit; a determination unit that determines a motion of the person, based on a detection result of the posture detection unit; and an instruction unit that generates an instruction that indicates a response to the person, based on a determination result of the determination unit. The monitoring apparatus can achieve highly functional monitoring in relation to operations on the money handling transaction apparatus.