Tire Size Identification Using Inner-Diameter Image Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating tire size in waste tire recycling are inaccurate, leading to significant revenue loss due to large margins of error in calculating tire size based on weight after fragmentation.
Innovation Solution
A tire-size identification method utilizing an object-detection model to frame tire images, followed by an image-segmentation model to determine the inner diameter, and a classification model to accurately classify tire sizes, combined with a tracking algorithm to follow tire positions and record maximum diameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If tire size is estimated by weighing tire fragments after chopping, then the operation process is simplified, but the measurement precision deteriorates significantly
Solution Approach 1:
The patent performs tire size measurement before the chopping process by capturing images of whole tires on the conveyor belt. The object detection model identifies and frames tire images, the image segmentation model extracts inner diameter information, and the classification model determines tire size categories before any fragmentation occurs, thus preserving measurement accuracy while maintaining operational simplicity
Solution Approach 2:
The patent replaces the mechanical weighing method with an optical-based computer vision system. Instead of physically chopping and weighing tire fragments, the system uses image capture, object detection, image segmentation, and classification algorithms to non-contactively measure tire inner diameter and determine size categories, eliminating the need for physical fragmentation while achieving precise measurement
2Measurement precision
If multiple processing steps (object detection, image segmentation, classification) are introduced to improve tire size identification accuracy, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent divides the tire size identification task into three distinct processing stages: object detection (framing tire images in the image), image segmentation (extracting tire inner diameter from framed images), and classification (categorizing tires by size based on inner diameter). Each stage is handled by a specialized model that focuses on a specific aspect of the measurement task, improving overall accuracy while keeping each individual component relatively simple and well-defined
3Productivity
If tire size is determined after fragmentation, then the workflow is streamlined, but revenue calculation accuracy deteriorates due to large margin of error
Solution Approach 1:
The patent performs tire size measurement and classification before the chopping process occurs in the recycling workflow. By capturing images and determining tire size categories while tires are still whole and positioned on the conveyor belt, the system establishes accurate size records that can be used for revenue calculation without being affected by subsequent fragmentation, thus maintaining both workflow efficiency and measurement accuracy
Data Source
AI summary
A tire-size identification method includes the following steps. An object-detection model frames a tire image in an image to generate a framed tire image. The framed tire image is input into an image-segmentation model. The image-segmentation model outputs the tire inner diameter of the tire. Moreover, the tire inner diameter is input into a classification model. The classification model outputs the tire size.