A Two-Stage Industrial Defect Detection Framework Based on Improved-YOLOv5 and Optimized-Inception-ResnetV2 Models
Aiming to address the currently low accuracy of domestic industrial defect detection, this paper proposes a Two-Stage Industrial Defect Detection Framework based on Improved-YOLOv5 and Optimized-Inception-ResnetV2, which completes positioning and classification tasks through two specific models. In...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/2/834 |
| _version_ | 1797496072793227264 |
|---|---|
| author | Zhuang Li Xincheng Tian Xin Liu Yan Liu Xiaorui Shi |
| author_facet | Zhuang Li Xincheng Tian Xin Liu Yan Liu Xiaorui Shi |
| author_sort | Zhuang Li |
| collection | DOAJ |
| description | Aiming to address the currently low accuracy of domestic industrial defect detection, this paper proposes a Two-Stage Industrial Defect Detection Framework based on Improved-YOLOv5 and Optimized-Inception-ResnetV2, which completes positioning and classification tasks through two specific models. In order to make the first-stage recognition more effective at locating insignificant small defects with high similarity on the steel surface, we improve YOLOv5 from the backbone network, the feature scales of the feature fusion layer, and the multiscale detection layer. In order to enable second-stage recognition to better extract defect features and achieve accurate classification, we embed the convolutional block attention module (CBAM) attention mechanism module into the Inception-ResnetV2 model, then optimize the network architecture and loss function of the accurate model. Based on the Pascal Visual Object Classes 2007 (VOC2007) dataset, the public dataset NEU-DET, and the optimized dataset Enriched-NEU-DET, we conducted multiple sets of comparative experiments on the Improved-YOLOv5 and Inception-ResnetV2. The testing results show that the improvement is obvious. In order to verify the superiority and adaptability of the two-stage framework, we first test based on the Enriched-NEU-DET dataset, and further use AUBO-i5 robot, Intel RealSense D435 camera, and other industrial steel equipment to build actual industrial scenes. In experiments, a two-stage framework achieves the best performance of 83.3% mean average precision (mAP), evaluated on the Enriched-NEU-DET dataset, and 91.0% on our built industrial defect environment. |
| first_indexed | 2024-03-10T01:58:26Z |
| format | Article |
| id | doaj.art-fdd2ea8d40854289ac0d96c8d48558d1 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T01:58:26Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-fdd2ea8d40854289ac0d96c8d48558d12023-11-23T12:53:39ZengMDPI AGApplied Sciences2076-34172022-01-0112283410.3390/app12020834A Two-Stage Industrial Defect Detection Framework Based on Improved-YOLOv5 and Optimized-Inception-ResnetV2 ModelsZhuang Li0Xincheng Tian1Xin Liu2Yan Liu3Xiaorui Shi4Center for Robotics, School of Control Science and Engineering, Shandong University, Jinan 250061, ChinaCenter for Robotics, School of Control Science and Engineering, Shandong University, Jinan 250061, ChinaCollege of Electronic Information and Optical Engineering, Nankai University, Tianjin 300350, ChinaCenter for Robotics, School of Control Science and Engineering, Shandong University, Jinan 250061, ChinaSinotruk Industry Park Zhangqiu, Sinotruk Jinan Power Co., Ltd., Jinan 250220, ChinaAiming to address the currently low accuracy of domestic industrial defect detection, this paper proposes a Two-Stage Industrial Defect Detection Framework based on Improved-YOLOv5 and Optimized-Inception-ResnetV2, which completes positioning and classification tasks through two specific models. In order to make the first-stage recognition more effective at locating insignificant small defects with high similarity on the steel surface, we improve YOLOv5 from the backbone network, the feature scales of the feature fusion layer, and the multiscale detection layer. In order to enable second-stage recognition to better extract defect features and achieve accurate classification, we embed the convolutional block attention module (CBAM) attention mechanism module into the Inception-ResnetV2 model, then optimize the network architecture and loss function of the accurate model. Based on the Pascal Visual Object Classes 2007 (VOC2007) dataset, the public dataset NEU-DET, and the optimized dataset Enriched-NEU-DET, we conducted multiple sets of comparative experiments on the Improved-YOLOv5 and Inception-ResnetV2. The testing results show that the improvement is obvious. In order to verify the superiority and adaptability of the two-stage framework, we first test based on the Enriched-NEU-DET dataset, and further use AUBO-i5 robot, Intel RealSense D435 camera, and other industrial steel equipment to build actual industrial scenes. In experiments, a two-stage framework achieves the best performance of 83.3% mean average precision (mAP), evaluated on the Enriched-NEU-DET dataset, and 91.0% on our built industrial defect environment.https://www.mdpi.com/2076-3417/12/2/834two-stagemultiscale detection layerEnriched-NEU-DETAUBO-i5 robot |
| spellingShingle | Zhuang Li Xincheng Tian Xin Liu Yan Liu Xiaorui Shi A Two-Stage Industrial Defect Detection Framework Based on Improved-YOLOv5 and Optimized-Inception-ResnetV2 Models Applied Sciences two-stage multiscale detection layer Enriched-NEU-DET AUBO-i5 robot |
| title | A Two-Stage Industrial Defect Detection Framework Based on Improved-YOLOv5 and Optimized-Inception-ResnetV2 Models |
| title_full | A Two-Stage Industrial Defect Detection Framework Based on Improved-YOLOv5 and Optimized-Inception-ResnetV2 Models |
| title_fullStr | A Two-Stage Industrial Defect Detection Framework Based on Improved-YOLOv5 and Optimized-Inception-ResnetV2 Models |
| title_full_unstemmed | A Two-Stage Industrial Defect Detection Framework Based on Improved-YOLOv5 and Optimized-Inception-ResnetV2 Models |
| title_short | A Two-Stage Industrial Defect Detection Framework Based on Improved-YOLOv5 and Optimized-Inception-ResnetV2 Models |
| title_sort | two stage industrial defect detection framework based on improved yolov5 and optimized inception resnetv2 models |
| topic | two-stage multiscale detection layer Enriched-NEU-DET AUBO-i5 robot |
| url | https://www.mdpi.com/2076-3417/12/2/834 |
| work_keys_str_mv | AT zhuangli atwostageindustrialdefectdetectionframeworkbasedonimprovedyolov5andoptimizedinceptionresnetv2models AT xinchengtian atwostageindustrialdefectdetectionframeworkbasedonimprovedyolov5andoptimizedinceptionresnetv2models AT xinliu atwostageindustrialdefectdetectionframeworkbasedonimprovedyolov5andoptimizedinceptionresnetv2models AT yanliu atwostageindustrialdefectdetectionframeworkbasedonimprovedyolov5andoptimizedinceptionresnetv2models AT xiaoruishi atwostageindustrialdefectdetectionframeworkbasedonimprovedyolov5andoptimizedinceptionresnetv2models AT zhuangli twostageindustrialdefectdetectionframeworkbasedonimprovedyolov5andoptimizedinceptionresnetv2models AT xinchengtian twostageindustrialdefectdetectionframeworkbasedonimprovedyolov5andoptimizedinceptionresnetv2models AT xinliu twostageindustrialdefectdetectionframeworkbasedonimprovedyolov5andoptimizedinceptionresnetv2models AT yanliu twostageindustrialdefectdetectionframeworkbasedonimprovedyolov5andoptimizedinceptionresnetv2models AT xiaoruishi twostageindustrialdefectdetectionframeworkbasedonimprovedyolov5andoptimizedinceptionresnetv2models |