Local Region Matching for Object Presence Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing monitoring systems struggle to efficiently detect and verify the presence or absence of objects, such as packages, over time, especially in environments with changing illumination conditions or shadows.
Innovation Solution
The system employs a computer-implemented method using a doorbell camera or related sensing device to detect human motion or robotic delivery vehicles. It generates multiple images, computes interest points through photometric augmentation, and uses anchor point estimation and bag-of-words descriptors for local region matching to monitor the presence or absence of objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional monitoring systems are used to detect objects, then the system can identify object presence, but it fails to accurately verify presence or absence over time, especially in changing illumination conditions
Solution Approach 1:
The system performs preliminary actions by capturing images at multiple time points (before and after motion detection) and pre-processing them through photometric augmentation to create augmented image sets. This allows the system to establish a baseline of what the region looks like under different lighting conditions before making final presence/absence determinations, thereby improving reliability without sacrificing precision under changing illumination.
Solution Approach 2:
The system applies photometric augmentation parameters (brightness, contrast, saturation adjustments) to create multiple versions of the same image under simulated different lighting conditions. By analyzing interest points across these parameter-changed images, the system can distinguish between actual object presence and apparent changes caused by illumination variations, resolving the contradiction between reliability and measurement precision.
2Productivity
If the system monitors objects over time to detect pickup or removal, then it provides comprehensive monitoring, but it requires processing multiple images which increases computational complexity
Solution Approach 1:
The system extracts only the necessary information by identifying and tracking interest points (local features) rather than processing the entire image. It extracts descriptors from these specific points and uses them for matching across time points. This extraction approach allows comprehensive temporal monitoring while reducing computational complexity by focusing processing only on salient regions rather than the complete image data.
Solution Approach 2:
The system segments the monitoring task into distinct stages: motion detection, image capture at multiple time points, interest point extraction, descriptor generation, and matching. By dividing the complex monitoring process into these manageable segments, the system achieves comprehensive temporal monitoring productivity while keeping each individual processing step computationally efficient and manageable.
3Measurement precision
If the system uses local region matching with anchor point estimation, then it can accurately track object presence, but it requires iterative computation of interest points which increases processing time
Solution Approach 1:
The system uses periodic action by iteratively computing interest points through multiple passes of photometric augmentation and interest point detection. Each iteration refines the interest point identification, progressively improving accuracy. This periodic refinement allows the system to achieve high measurement precision in local region matching while the iterative nature provides a systematic approach that converges on accurate results without excessive processing time.
Data Source
AI summary
Methods and systems, including computer-readable media, are described for monitoring presence or absence of an object at a property using local region matching. A system generates images while monitoring an area of the property and, based on the images, detects an object in a region of interest in the area. For each of the images: the system iteratively computes interest points for the object using photometric augmentation applied to the image before each iteration of computing the interest points. A digital representation of the region and the object is generated based on interest points that repeat across the images after each application of the photometric augmentation. Based on the digital representation, a set of anchor points are determined from the interest points that repeat across images. Using the set of anchor points, the system detects an absence or a continued presence of the object in the area of the property.