Survey of Quantization Techniques for On-Device Vision-based Crack Detection

2025-02-04

AI Summary

In edge-based structural health monitoring using unmanned aerial vehicles (UAVs), there exists a fundamental trade-off between stringent resource constraints of embedded devices and the requirements for real-time, low-power crack detection. Method: This study systematically evaluates dynamic quantization, post-training quantization (PTQ), and quantization-aware training (QAT) across TensorFlow, PyTorch, and ONNX frameworks, applied to lightweight CNNs—MobileNetV1×0.25 and MobileNetV2×0.5—for accuracy–efficiency trade-off analysis. Contribution/Results: We demonstrate, for the first time in edge vision detection, that QAT achieves near-full-precision performance (F1 = 0.8376), substantially outperforming PTQ (prone to accuracy instability) and dynamic quantization (limited PyTorch deployment support). The proposed Torch-QAT–enhanced MobileNetV2×0.5 configuration delivers high accuracy, low memory footprint, and cross-platform deployability. Our work establishes an empirical benchmark and practical quantization pathway for deploying lightweight CNNs efficiently under severe resource constraints.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Learning & Optimization for CV
• Knowledge Representation and Reasoning: Qualitative Reasoning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Structural Health Monitoring (SHM) ensures the safety and longevity of infrastructure by enabling timely damage detection. Vision-based crack detection, combined with UAVs, addresses the limitations of traditional sensor-based SHM methods but requires the deployment of efficient deep learning models on resource-constrained devices. This study evaluates two lightweight convolutional neural network models, MobileNetV1x0.25 and MobileNetV2x0.5, across TensorFlow, PyTorch, and Open Neural Network Exchange platforms using three quantization techniques: dynamic quantization, post-training quantization (PTQ), and quantization-aware training (QAT). Results show that QAT consistently achieves near-floating-point accuracy, such as an F1-score of 0.8376 for MBNV2x0.5 with Torch-QAT, while maintaining efficient resource usage. PTQ significantly reduces memory and energy consumption but suffers from accuracy loss, particularly in TensorFlow. Dynamic quantization preserves accuracy but faces deployment challenges on PyTorch. By leveraging QAT, this work enables real-time, low-power crack detection on UAVs, enhancing safety, scalability, and cost-efficiency in SHM applications, while providing insights into balancing accuracy and efficiency across different platforms for autonomous inspections.

Problem

• Optimize deep learning for crack detection
• Evaluate quantization techniques on neural networks
• Enable real-time detection on resource-constrained devices

Innovation

• Quantization-aware training enhances model accuracy
• Post-training quantization reduces memory and energy use
• Dynamic quantization maintains accuracy with deployment challenges