Multimodal Companion Animal Identification for Lost-Pet Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing companion animal identification systems face challenges in accuracy due to issues with face recognition, nose print recognition, and voice recognition, particularly in managing companion animals and tracking them in case of loss, with existing methods lacking reliability and efficiency.
Innovation Solution
A system that utilizes face recognition, nose print recognition, and motion recognition, employing artificial neural networks to analyze video or image data, and applying weights to determination information for improved identification, including normalization and histogram modeling to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple recognition methods (face, nose print, voice, motion) are integrated to improve identification reliability, then identification accuracy is improved, but device complexity increases
Solution Approach 1:
The patent combines multiple recognition methods (face recognition, nose print recognition, voice recognition, and motion recognition) into a single integrated identification system. The server receives data from all four recognition units and processes them together to generate comprehensive identification results, thereby improving reliability through multi-modal verification while managing system complexity through unified architecture.
Solution Approach 2:
The identification server is designed to handle multiple types of recognition data simultaneously. It can process face data, nose print data, voice data, and motion data through a single unified processing platform, making the system multi-functional and versatile in handling different biometric modalities without requiring separate dedicated systems for each recognition type.
2Ease of manufacture
If face recognition is performed using general methods trained on human faces, then implementation is straightforward, but identification accuracy for animals decreases
Solution Approach 1:
The patent applies parameter changes by adapting face recognition algorithms specifically for animal faces rather than using human-face-trained models. This involves modifying the feature extraction parameters, training data characteristics, and recognition thresholds to match animal facial structures, thereby improving identification accuracy while maintaining implementation feasibility through adapted rather than completely redesigned systems.
3Device complexity
If only general feature point extraction is used for nose print recognition, then processing is simple, but recognition accuracy is insufficient for commercialization
Solution Approach 1:
The patent segments the nose print analysis into distinct processing stages: first extracting key feature points from the nose region, then performing contrast-limited adaptive histogram equalization to enhance local contrast, and finally applying scale-invariant feature transform for robust feature description. This segmentation allows each processing step to be optimized independently, achieving high recognition accuracy while keeping individual processing modules manageable in complexity.
Solution Approach 2:
The patent performs preliminary actions by pre-processing the nose print images through contrast-limited adaptive histogram equalization before feature extraction. This preliminary enhancement of image contrast and quality ensures that subsequent feature point extraction and matching operations work with optimized input data, thereby improving final recognition accuracy without significantly increasing overall system complexity.
Data Source
AI summary
The present invention relates to a technology capable of identifying a companion animal by analyzing video, image, and voice data collected through CCTV or a camera, for management of companion animals and tracking in case of loss thereof, wherein, by simultaneously or sequentially using at least one identification method among facial recognition, nose print recognition, voice recognition, and motion recognition by analyzing a video, image, or voice, an effect of greatly improving the reliability of object identification for companion animals can be provided.