Smart Defect Inspection: Portwell × Neurocle’s AI Inspection Solution for Automotive Parts and Tires

A New Opportunity for Smart Manufacturing Powered by Edge AI

In smart manufacturing and automotive inspection, defect inspection driven by Edge AI and Machine Vision has become a standardized trend for quality assurance. The global automotive AI inspection system market was approximately USD 4–5 billion in 2024 and is projected to expand at a compound annual growth rate (CAGR) of 40–45% from 2025 to 2034.

The tire X-ray inspection market, driven by vehicle safety and performance requirements, reached about USD 200–250 million in 2023 and is expected to grow to USD 300 million between 2024 and 2032. Faced with key trends like smart manufacturing, edge computing, and AIoT, manufacturers must combine advanced algorithms with high-performance computing platforms to meet real-time, high-accuracy, and scalable inspection demands.

The Dual Challenge of Diverse Defects and Environmental Variability

In real-world applications of automotive parts and tire inspection, the main challenges can be summarized as:

1. Variety of defect types: Defects such as bolt deformation, cracks, and internal tire air bubbles impose different requirements on image resolution, lighting adaptation, and post-processing capabilities.
2. Real-time, low latency: As production line speeds increase, the Edge AI system must complete image capture, preprocessing, inference, and feedback within milliseconds; otherwise, line efficiency and downstream logistics scheduling will be affected.
3. Environmental variability: Workshop lighting changes, reflective surfaces, high temperatures, and high humidity can all cause Machine Vision algorithms to misjudge or miss defects, undermining quality control.
4. Big data training and model deployment: Building highly accurate anomaly detection models requires large amounts of annotated data and continuous optimization; deploying these models on-site must balance computing power and energy consumption.

An Integrated AIoT Architecture

To address the above challenges, Portwell and partner Neurocle offer a complete AIoT solution:

1. Using the no-code, GUI-based NEURO-T Model Trainer, users can build high-performance inspection models within hours via Synthetic Defect Generation and Unsupervised Anomaly training modes. Dynamic tuning of learning rate, model architecture, and optimizer parameters ensures optimal inference performance on edge devices.
2. The highly optimized, deep–learning–based NEURO-R Runtime Library is then deployed on the PJAI-100 for inference. It supports nine inspection model types, including real-time object detection, classification, and semantic segmentation, to achieve on-site smart inspection and data feedback.

High-Performance PJAI-100 Edge Computing System

The PJAI-100, built on NVIDIA Jetson Orin NX/Nano modules, delivers up to 100 TOPS of computing performance. It features an M.2 2280 M-Key NVMe (PCIe x4), an M.2 2230 E-Key, and a Nano-SIM slot, supporting architectures for remote monitoring and edge-side data uplink systems.

Integration Advantages: Precision, Low Latency, and Scalable Deployment

By integrating the above software and hardware, customers can achieve:

1. 99.9% inspection precision for bolt and tire defects
2. Millisecond-level inference latency
3. Scalable multi-camera deployment
4. This reduces yield losses and enables data-driven optimization of production processes, improving overall equipment effectiveness and asset utilization.

Towards a New Era of Smart Quality Management

As the automotive parts inspection and tire X-ray markets face dual demands for line automation and intelligent quality control, the Portwell PJAI-100, combined with the Neurocle solution, provides an integrated Edge AI Machine Vision platform. It delivers low-latency inference, high accuracy, and scalable deployment, making it the ideal choice for automakers seeking high yield management and parts suppliers requiring real-time inspection. Enterprises can realize smart manufacturing and quality management by leveraging Portwell’s R&D and manufacturing expertise, global distribution channels, and local services.