An Improved Mask R-CNN Micro-Crack Detection Model for the Surface of Metal Structural Parts
Micro-crack detection is an essential task in critical equipment health monitoring. Accurate and timely detection of micro-cracks can ensure the healthy and stable service of equipment. Aiming at improving the low accuracy of the conventional target detection model during the task of detecting micro...
| 主要な著者: | , , , , |
|---|---|
| フォーマット: | 論文 |
| 言語: | English |
| 出版事項: |
MDPI AG
2023-12-01
|
| シリーズ: | Sensors |
| 主題: | |
| オンライン･アクセス: | https://www.mdpi.com/1424-8220/24/1/62 |
| _version_ | 1827384485606326272 |
|---|---|
| author | Fan Yang Junzhou Huo Zhang Cheng Hao Chen Yiting Shi |
| author_facet | Fan Yang Junzhou Huo Zhang Cheng Hao Chen Yiting Shi |
| author_sort | Fan Yang |
| collection | DOAJ |
| description | Micro-crack detection is an essential task in critical equipment health monitoring. Accurate and timely detection of micro-cracks can ensure the healthy and stable service of equipment. Aiming at improving the low accuracy of the conventional target detection model during the task of detecting micro-cracks on the surface of metal structural parts, this paper built a micro-cracks dataset and explored a detection performance optimization method based on Mask R-CNN. Firstly, we improved the original FPN structure, adding a bottom-up feature fusion path to enhance the information utilization rate of the underlying feature layer. Secondly, we added the methods of deformable convolution kernel and attention mechanism to ResNet, which can improve the efficiency of feature extraction. Lastly, we modified the original loss function to optimize the network training effect and model convergence rate. The ablation comparison experiments shows that all the improvement schemes proposed in this paper have improved the performance of the original Mask R-CNN. The integration of all the improvement schemes can produce the most significant performance improvement effects in recognition, classification, and positioning simultaneously, thus proving the rationality and feasibility of the improved scheme in this paper. |
| first_indexed | 2024-03-08T14:58:31Z |
| format | Article |
| id | doaj.art-91597b778a034a3d9a1bd4091ed62795 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-08T14:58:31Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-91597b778a034a3d9a1bd4091ed627952024-01-10T15:08:24ZengMDPI AGSensors1424-82202023-12-012416210.3390/s24010062An Improved Mask R-CNN Micro-Crack Detection Model for the Surface of Metal Structural PartsFan Yang0Junzhou Huo1Zhang Cheng2Hao Chen3Yiting Shi4School of Mechanical Engineering, Dalian University of Technology, Dalian 116024, ChinaSchool of Mechanical Engineering, Dalian University of Technology, Dalian 116024, ChinaSchool of Mechanical Engineering, Dalian University of Technology, Dalian 116024, ChinaSchool of Mechanical Engineering, Dalian University of Technology, Dalian 116024, ChinaSchool of Mechanical Engineering, Dalian University of Technology, Dalian 116024, ChinaMicro-crack detection is an essential task in critical equipment health monitoring. Accurate and timely detection of micro-cracks can ensure the healthy and stable service of equipment. Aiming at improving the low accuracy of the conventional target detection model during the task of detecting micro-cracks on the surface of metal structural parts, this paper built a micro-cracks dataset and explored a detection performance optimization method based on Mask R-CNN. Firstly, we improved the original FPN structure, adding a bottom-up feature fusion path to enhance the information utilization rate of the underlying feature layer. Secondly, we added the methods of deformable convolution kernel and attention mechanism to ResNet, which can improve the efficiency of feature extraction. Lastly, we modified the original loss function to optimize the network training effect and model convergence rate. The ablation comparison experiments shows that all the improvement schemes proposed in this paper have improved the performance of the original Mask R-CNN. The integration of all the improvement schemes can produce the most significant performance improvement effects in recognition, classification, and positioning simultaneously, thus proving the rationality and feasibility of the improved scheme in this paper.https://www.mdpi.com/1424-8220/24/1/62mask R-CNNmicro-cracktarget detectionmetal structural partsdeformable convolution kernelattention mechanism |
| spellingShingle | Fan Yang Junzhou Huo Zhang Cheng Hao Chen Yiting Shi An Improved Mask R-CNN Micro-Crack Detection Model for the Surface of Metal Structural Parts Sensors mask R-CNN micro-crack target detection metal structural parts deformable convolution kernel attention mechanism |
| title | An Improved Mask R-CNN Micro-Crack Detection Model for the Surface of Metal Structural Parts |
| title_full | An Improved Mask R-CNN Micro-Crack Detection Model for the Surface of Metal Structural Parts |
| title_fullStr | An Improved Mask R-CNN Micro-Crack Detection Model for the Surface of Metal Structural Parts |
| title_full_unstemmed | An Improved Mask R-CNN Micro-Crack Detection Model for the Surface of Metal Structural Parts |
| title_short | An Improved Mask R-CNN Micro-Crack Detection Model for the Surface of Metal Structural Parts |
| title_sort | improved mask r cnn micro crack detection model for the surface of metal structural parts |
| topic | mask R-CNN micro-crack target detection metal structural parts deformable convolution kernel attention mechanism |
| url | https://www.mdpi.com/1424-8220/24/1/62 |
| work_keys_str_mv | AT fanyang animprovedmaskrcnnmicrocrackdetectionmodelforthesurfaceofmetalstructuralparts AT junzhouhuo animprovedmaskrcnnmicrocrackdetectionmodelforthesurfaceofmetalstructuralparts AT zhangcheng animprovedmaskrcnnmicrocrackdetectionmodelforthesurfaceofmetalstructuralparts AT haochen animprovedmaskrcnnmicrocrackdetectionmodelforthesurfaceofmetalstructuralparts AT yitingshi animprovedmaskrcnnmicrocrackdetectionmodelforthesurfaceofmetalstructuralparts AT fanyang improvedmaskrcnnmicrocrackdetectionmodelforthesurfaceofmetalstructuralparts AT junzhouhuo improvedmaskrcnnmicrocrackdetectionmodelforthesurfaceofmetalstructuralparts AT zhangcheng improvedmaskrcnnmicrocrackdetectionmodelforthesurfaceofmetalstructuralparts AT haochen improvedmaskrcnnmicrocrackdetectionmodelforthesurfaceofmetalstructuralparts AT yitingshi improvedmaskrcnnmicrocrackdetectionmodelforthesurfaceofmetalstructuralparts |