BRAV-7120 Powers AI Machine Vision: Achieving High-Precision Appearance Inspection on a Spanish Automobile Wheel Production Line

Driven by Industry 4.0 and smart manufacturing, machine vision technology is becoming a core tool for quality control in modern manufacturing. In particular, the automotive parts industry demands extremely high precision in product appearance and dimensional accuracy.

As a key component of a vehicle’s driving system, the automotive wheel hub directly affects driving safety and user experience through its surface quality, dimensional precision, and structural integrity. Traditional manual inspection methods are inefficient, prone to fatigue, and highly subjective, making them difficult to meet the demands of large-scale, highly consistent production. The AI-based machine vision system can achieve automated, non-contact, high-precision real-time detection, greatly improving production efficiency and quality control levels.

Customer Application Requirements

A Spanish automotive wheel manufacturer sought to introduce an intelligent vision inspection system to achieve the following objectives:

• Defect Detection: Automatically identify surface defects such as scratches, cracks, shrinkage, and paint peeling on the wheel hub
• Dimension Measurement: Accurately measure key dimensions such as the wheel hub diameter, bolt hole position, and eccentricity.
• Real-Time Processing: Achieve millisecond-level response time on the production line, supporting real-time sorting and rejection of defective products.
• System Integration: Support for simultaneous multi-camera acquisition and compatibility with 2D/3D vision sensors and lighting systems.

To meet the intelligent manufacturing needs of the automotive wheel production line, the customer proposed the following core requirements for the AI vision controller:

⭐ High Performance: Equipped with AI computing power, it supports real-time processing of multi-channel camera data and runs deep learning models.

⭐ Multi-Camera Access: Supports connection to multiple 2D/3D industrial cameras for stable high-definition image capture.

⭐ Industrial-Grade Reliability: Featuring a fanless design, it supports a wide 12-24V input voltage and operates in a -20°C to 60°C operating temperature range.

⭐ Extensive Industrial Interfaces: Provides isolated I/O, CAN, and serial ports for easy integration with PLCs, robots, and other equipment.

⭐ Continuous and Stable Operation: Enables 24/7 uninterrupted operation, improving production efficiency.

JHCTECH Solutions

After comprehensive testing, JHCTECH’s BRAV-7120 embedded AI edge computing system was selected as the ideal solution for intelligent appearance inspection on the automotive wheel production line.

The BRAV-7120, built on the NVIDIA Jetson Orin Nano platform, boasts powerful AI computing power ranging from 20 to 67 TOPS. It offers a rich set of interfaces (2x LAN, 4x USB, 2x COM, and 2x CAN), allowing it to connect two to three 2D/3D cameras, sensors, lighting, and optical components, depending on the project. It captures high-definition images in real time and performs intelligent analysis. By integrating advanced deep learning algorithms, the system accurately identifies various defects on the wheel hub surface, including scratches, cracks, shrinkage cavities, and paint peeling, while simultaneously measuring key dimensions and determining errors, effectively improving inspection accuracy and consistency.

In addition, the BRAV-7120 supports four-channel GMSL or MIPI CSI camera expansion, multiple isolated I/O and communication interfaces, allowing flexible connection to sensors, PLCs, and peripheral actuators to meet the complex integration demands of industrial environments.

The device also features a DC 9-36V wide-voltage input, fanless operation in a wide operating temperature range (-20°C to 60°C), and multiple isolation protections. This adapts to voltage fluctuations and electromagnetic interference in production environments, ensuring 24/7 continuous and stable system operation.

Combined with the customer’s pre-installed AI vision inspection software, it enables end-to-end low-latency processing from image acquisition, model inference, to output results. It can also be seamlessly integrated into the production line's MES or quality management platform, ultimately helping customers achieve a highly efficient, reliable, and fully automated wheel quality inspection and sorting process.

The JHCTECH BRAV-7120, with its powerful AI computing power, rich interface resources, and industrial-grade reliability, provides a cost-effective and highly reliable machine vision solution for the automotive wheel manufacturing industry. In the future, JHCTECH will continue to delve into smart manufacturing, smart transportation, logistics testing and other fields to support more companies in achieving intelligent upgrades.

JHCTECH — Intelligent Edge, Driving the Future.