FedTR: Federated Learning Framework with Transfer Learning for Industrial Visual Inspection

2026-07-08

AI Summary

This work addresses the performance degradation of federated learning in industrial visual inspection caused by data scarcity and heterogeneity. To this end, the authors propose FedTR, a novel framework that effectively integrates transfer learning into the federated learning pipeline for the first time. FedTR first pretrains a model on publicly available data and then fine-tunes it on distributed private data under privacy-preserving constraints, enabling end-to-end text recognition for label defect detection. The approach substantially mitigates training challenges arising from limited sample sizes and non-identical data distributions. Evaluated on a cartridge label defect detection task, FedTR achieves word-level recognition accuracies of 95.5% and 94.2% under homogeneous and heterogeneous settings, respectively, matching the performance of centralized training.

Abstract

Federated learning (FL) is a collaborative learning scheme to train deep learning models, where collaborating parties can consolidate their models without sharing local data with other parties, hence preserving data privacy. Nevertheless, when implementing FL in Industrial visual inspection (IVI), the constraints posed by limited data availability and the intricate nature of the inspection tasks significantly impact the performance of the resulting model. This paper introduces FedTR, a novel FL framework incorporating transfer learning designed for Autonomous IVI, focusing on the challenging task of identifying label defects through end-to-end text recognition. Transfer learning is a method that leverages the knowledge of a pre-trained model to adapt to a different dataset. FedTR initially trains the model using a publicly available dataset, after which performs the essential federated learning process with model fine-tuning on the distributed and limited private data. Extensive experiment results demonstrate the effectiveness and feasibility of FedTR on private ink cartridge datasets for label defect identification. FedTR achieves an end-to-end text recognition word-level accuracy of 95.5% and 94.2% on homogeneous and heterogeneous data respectively. Additionally, it attains performance levels that are on par with those achieved through centralized training.

Problem

• Federated Learning
• Industrial Visual Inspection
• Transfer Learning
• Label Defect Identification
• Data Scarcity

Innovation

• Federated Learning
• Transfer Learning
• Industrial Visual Inspection
• Label Defect Detection
• End-to-End Text Recognition