Far-Infrared Recognition Device for Human-Animal Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing safety support systems for vehicles struggle to accurately distinguish between humans and animals when they appear in images showing their front or back, leading to false detection.
Innovation Solution
A recognition apparatus and method utilizing a far-infrared camera to acquire images and determine whether an object is a human or an animal based on temperature variations within a target region, employing recognition dictionaries and image processing techniques to differentiate between human and animal characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If recognition is based on object characteristics (size, shape, position) when animal faces sideways, then detection accuracy is improved, but false detection occurs when human or animal faces front or back
Solution Approach 1:
The patent changes the detection parameter from geometric characteristics (size, shape, position) to thermal characteristics (temperature distribution, temperature difference between regions). This allows accurate identification of humans vs. animals regardless of their facing direction, as thermal patterns differ fundamentally between species even when geometric patterns become similar.
Solution Approach 2:
The patent introduces temperature distribution as an intermediary characteristic for identification. Instead of directly using geometric features that fail when facing direction changes, the system uses thermal radiation patterns as a mediating property that maintains discriminative power across different orientations.
2Ease of operation
If geometric characteristics are used for identification, then detection is simple when facing sideways, but identification becomes difficult when facing front or back
Solution Approach 1:
The patent transitions from using geometric parameters (simple when facing sideways) to thermal parameters (consistently distinguishable regardless of orientation). Temperature distribution provides a new parameter space where identification remains simple and reliable across all facing directions.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively reduces false detection between humans and animals by accurately determining their presence and type, even when they are facing front or back, enhancing safety by providing accurate warnings to drivers.
Implementation Method 1
acquire an image captured by a far-infrared camera
Data Source
AI summary
A recognition apparatus (20) includes an image acquisition unit (21) configured to acquire an image captured by a far-infrared camera (10) and a recognition processing unit (22) configured to set a target region in the image and determine whether an object in the target region is a human being or an animal based on a temperature variation in the target region. A recognition apparatus, a recognition method, and a recognition program that can correctly determine whether it is a human being or an animal that appears in an image even when the human being or the animal appears in the image showing his/her or its front or back can be provided.