Monitoring Camera Image Identification for Leaving Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional monitoring cameras waste computation time and fail to provide accurate results when identifying whether an object is a leaving or missing object due to detecting irrelevant changes such as deformation, movement, or intensity variations caused by weather, wind, or accidents, which are not of interest to the user.
Innovation Solution
An image identifying method that acquires a foreground region corresponding to the target object, analyzes if it conforms to a variant feature, and compares it with a reference image to determine if the object is leaving or missing, using techniques like convolutional neural networks and feature detection to filter out irrelevant changes, thereby reducing computational load and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional monitoring camera compares current monitoring image with previous background image for detecting pixel variation, then the monitoring camera can detect object changes such as deformation, movement, or intensity variation, but the monitoring camera cannot accurately identify whether the object is a leaving object or a missing object due to irrelevant changes
Solution Approach 1:
The patent segments the image processing into distinct stages: first detecting pixel variation to identify potential changes, then analyzing whether changes conform to variant features (deformation, movement, intensity variation), and finally comparing with reference images only for non-variant objects. This segmentation allows the system to filter out irrelevant changes before final identification, improving both accuracy and reliability.
Solution Approach 2:
The patent extracts and analyzes specific feature types (deformation, movement, intensity variation) separately from the overall image comparison process. By taking out these variant features and handling them differently from non-variant objects, the system can focus computational resources on accurate identification of leaving and missing objects while ignoring irrelevant changes.
2Measurement precision
If conventional monitoring camera executes object identifying process to detect all pixel variations, then the monitoring camera can detect various object changes, but the monitoring camera wastes computation time and cannot provide accurate identifying result
Solution Approach 1:
The patent performs preliminary analysis by first detecting pixel variation and then determining whether the variation conforms to known variant features (deformation, movement, intensity variation) before proceeding to the computationally intensive object identification process. This preliminary filtering action eliminates irrelevant objects from further processing, significantly reducing computation time while maintaining identification accuracy for relevant objects.
Solution Approach 2:
The patent applies partial action by not performing full object identification on all detected variations. Instead, it selectively applies identification only to objects that do not conform to variant features, while using simplified variant feature analysis for objects that do conform. This partial application of the full identification process optimizes computation time while maintaining accuracy.
3Adaptability or versatility
If conventional monitoring camera detects all pixel variations including deformation, movement, and intensity variation, then the monitoring camera can comprehensively monitor object changes, but the monitoring camera cannot distinguish relevant changes from irrelevant changes
Solution Approach 1:
The patent applies different processing qualities to different types of detected variations. Variant features (deformation, movement, intensity variation) receive localized analysis focused on their specific characteristics, while non-variant objects receive full object identification processing. This local quality differentiation allows comprehensive monitoring of all change types while maintaining high identification precision for relevant objects.
Data Source
AI summary
An image identifying method is applied to a monitoring camera and a monitoring camera system and used to determine whether a target object is a leaving object or a missing object. The image identifying method includes acquiring a foreground region within a monitoring image corresponding to the target object, analyzing whether the target object inside the foreground region conforms to a variant feature, and comparing the foreground region with a reference image for determining the target object belongs to the leaving object or the missing object when the foreground region does not conform to the variant feature.