Cross-Scenario Industrial Defect Detection via Domain-Robust Training and Metric-Aligned Selection

Dikai Li; Weijia Wang; Yuan-Gen Wang

cs.CVOct 14, 2026

Cross-Scenario Industrial Defect Detection via Domain-Robust Training and Metric-Aligned Selection

Authors: Dikai Li, Weijia Wang, Yuan-Gen Wang

Organizations: School of Artificial Intelligence, Guangzhou University, Guangzhou 510006, China

Abstract

Automated defect detection of industrial boards must remain highly reliable under cross-scenario variations in lighting, surface contaminants, and production line conditions. This makes cross-scenario defect detection a domain generalization problem, wherein models are required to detect, localize, and classify defects across unseen scenarios while minimizing false alarms on clean boards. For the IDA 2026 Track 1 challenge, we propose a cross-scenario defect detection method that includes four complementary modules: cross-scenario robust training, small-defect-aware learning, score calibration and inference threshold selection, and metric-aligned checkpoint selection. These four modules respectively introduce cross-scenario robust normalization and DSU-style statistic perturbation with progressive random convolution, NWD and parameter-free spatial attention, the VFL learning objective, and a custom SWAD variant driven by the Track 1-oriented proxy. Experimental results demonstrate that our model achieves a Track 1 score of 0.564 with a screening score of 0.696 on the test dataset on a single RTX 4090 GPU. The code is available at https://github.com/GZHU-DVL/Dikai.

Figures & tables

Fig. 1: Overview of the proposed method for cross-scenario industrial defect detection.

Evaluation	$S_{\text{loc}}$	$S_{\text{cls}}$	$S_{\text{screen}}$	$S_{\text{Track1}}$
Local official	0.826	0.436	0.543	0.596
Challenge test feedback	0.489	0.465	0.696	0.564
$\Delta$ (local $\to$ official test)	$-0.337$	$+0.029$	$+0.153$	$-0.032$

TABLE I: Local re-evaluation and official challenge test feedback for the queried long-schedule checkpoint.

Short schedule (600 ep.)
Variant	$S_{\text{loc}}$	$S_{\text{cls}}$	$S_{\text{screen}}$	$S_{\text{Track1}}$
Full recipe	0.819	0.451	0.629	0.632
no MixRand	0.818	0.450	0.624	0.630
MixRand, no attenuation	0.821	0.453	0.622	0.631
no SWAD avg.	0.817	0.428	0.602	0.614
2000-ep queried reference (local re-eval.)

TABLE II: Targeted local ablation under a unified 600-epoch protocol. The queried 2000-epoch checkpoint is included as a local re-evaluation reference only.

Fig. 2: Local Track 1 component scores for the comparable 600-epoch ablations. The no-SWAD variant shows the largest drop on

S_{\text{screen}}

and

S_{\text{cls}}

, whereas the MixRand variants remain close to the full recipe.

Fig. 3: Training dynamics of the full recipe. Solid: smoothed Track 1 proxy

F_{\text{T1}}

(Eq. 1 ). Dashed:

P_{\text{screen}}

component alone. Dotted: default mAP-based detector fitness (

0.1\cdot

mAP50

+0.9\cdot

mAP50-95). The shaded interval is the SWAD averaging window selected from the screening-weighted proxy rather than from the default detector fitness.

Fig. 4: Qualitative predictions on the local validation split. Top row: full recipe with MixRand disabled (the “no_mixrand” ablation). Bottom row: full recipe. Columns show selected clean and defect-present cases from the local validation split. The figure is illustrative; quantitative claims are reported in Table II .

Industrial Anomaly Detection

Industrial anomaly detection finds defective products and abnormal equipment behavior from visual, sensor, audio, and other production data.

See all 260 papers

ICME 2026 Grand Challenge on Cross-Scenario Defect Detection and Fine-Grained Severity Grading for High-Precision Manufacturing
Jul 6, 2026Wei Sun, Weixia Zhang, Linhan Cao +30Industrial Anomaly Detection Industrial Inspection
East China Normal University, Shanghai Jiao Tong University, Tsinghua Shenzhen International Graduate School, and Universit´e de Nantes.
Feature Perturbation Pool-based Fusion Network for Unified Multi-Class Industrial Defect Detection
Apr 21, 2026Yuanchan Xu, Wenjun Zang, Ying WuIndustrial Anomaly Detection
School of Information Engineering, Sichuan University, China
Continuity-Driven Representation Learning for Industrial Defect Detection
Aug 18, 2026Minjong Kim, Hyun Jun Kim, Jeongrae Kim +2Object Detection Industrial Anomaly Detection
Chung-Ang University Seoul, Republic of Korea

Jul 6, 2026J/K move · Enter open · S save

ICME 2026 Grand Challenge on Cross-Scenario Defect Detection and Fine-Grained Severity Grading for High-Precision Manufacturing

Wei Sun, Weixia Zhang, Linhan Cao, Mingkai Lu+29

East China Normal University, Shanghai Jiao Tong University, Tsinghua Shenzhen International Graduate School, and Universit´e de Nantes.

Industrial Anomaly Detection Industrial Inspection Cross-Domain Generalization Automated Grading

Open paper AskPDF arXiv

This paper presents the IEEE International Conference on Multimedia and Expo (ICME) 2026 Grand Challenge on Cross-Scenario Defect Detection and Fine-Grained Severity Grading for High-Precision Manufacturing. The challenge is motivated by two key limitations of existing industrial defect inspection systems: (1) current deep learning-based methods often suffer significant performance degradation when deployed in unseen production scenarios, and (2) most benchmarks neglect severity-aware assessment, which is critical for risk control and yield optimization. To address these limitations, we design two complementary tracks: Track 1 (Cross-Scenario Defect Detection) targets accurate defect detection, localization, and classification across diverse unseen production environments; Track 2 (Fine-Grained Severity Grading) requires assigning each detected defect an industry-standard severity level, including Acceptable, Marginal NG, NG, and Gross NG. We construct a large-scale industrial dataset of high-resolution microscopic images spanning seven representative defect categories, comprising over 3,800 images with pixel-level instance annotations for Track 1 and over 2,600 images with severity-grade labels for Track 2. The challenge attracted 86 registered participants with 130 submissions; during the final testing phase, 21 teams submitted results and 12 teams provided models with technical reports. The resulting benchmark, together with the diverse and effective solutions contributed by participating teams, sets a new standard for industrial defect analysis research.

Abstract

Industrial Anomaly Detection

Explore similar work

ICME 2026 Grand Challenge on Cross-Scenario Defect Detection and Fine-Grained Severity Grading for High-Precision Manufacturing

Feature Perturbation Pool-based Fusion Network for Unified Multi-Class Industrial Defect Detection

Continuity-Driven Representation Learning for Industrial Defect Detection