System functions
The used tire sorting machine integrates into the customer’s warehouse to automate the identification, sorting, and tracking processes:
- Scans the data on the tire sidewall
- Performs 3D measurements of the tread
- Takes a photo for e-commerce
- Prints the label for the warehouse
The 3D vision system integrated in the machine was developed using artificial intelligence and integrated into the fully automatic conveyor belt machine to detect—without handling the tire—the brand, model, diameter, load index, DOT number, and much more, printed on the tire in any position.
The system operates without prior knowledge of the tires’ diameters and heights; it only requires that the tires be loaded onto the conveyor belt. Even if loaded in contact with one another, the machine separates them one by one to ensure perfect reading.
The 3D image of the tire is reconstructed using laser line scanning and programmable illuminators to read the raised characters and assess tire wear through objective measurements across a large sampling area.
Finally, a labeling station prints the part information detected by the machine vision system, and the option of capturing a high-resolution image with background removal, ready for e-commerce.
Key features
The main features of the machine are:
Realization
When a tire enters the cabin, the vision system scans it and extracts the embossed markings on the sidewall, where the DOT code and tire specifications are located.
This innovative reading technique makes it possible to highlight the embossed characters on the tire’s sidewall, despite the challenge of capturing high-contrast images of a completely black material like a tire.
When inspecting tread wear, the measurement is performed using a 3D laser scan, which is capable of mapping the geometry of the tire’s tread pattern. The vision system analyzes the resulting three-dimensional model, calculating the remaining tread depth measured in the main grooves between the highest and lowest points from the groove to the tire’s tread blocks. This inspection takes place as the tire passes through the inspection area (without rotation).
The proposed vision system requires no part handling and can operate from either the bottom or the top, depending on the configuration.
It allows you to classify different formats loaded onto the conveyor belt without needing to know in advance the dimensions and model of each tire.
Data reading and wear assessment.
The 3D image of the tire is reconstructed using laser line scanning and programmable illuminators to read the raised markings and quantify the degree of tread wear.
The process of reading the information and selecting which data to store is based on deep learning with a hybrid solution that combines the strengths of neural networks with traditional algorithms: reliability, accuracy and generalization.
Instant label with tire data
All the characteristic data for each tire —which the system has automatically read— is sent to the label printer for product tracking.
To generate and print complete identification labels, we recommend using an industrial printer capable of ensuring high reliability, speed, and print quality. The basic solution, which includes a manual label applicator, is supplied ready for integration with an automatic applicator.
Each label contains all the information and technical specifications previously collected during the tire inspection. It may include a barcode or QR code to enable quick and accurate identification during logistics and warehouse operations.
High-definition photos
The image capture station is designed to capture high-resolution images of the tire as viewed from the tread side. It uses 2D industrial color cameras positioned laterally along the tire’s travel path to ensure uniformity, repeatability, and consistent quality in image capture. The background is removed for publication on the company’s website, in digital price lists, and online catalogs, providing the end customer with a clear and comprehensive view of the tire’s aesthetic condition and visual characteristics.
Quick calibration
Calibration does not require skilled personnel because it follows a simple monitor-guided procedure.
