A Novel Method for Filled/Unfilled Grain Classification Based on Structured Light Imaging and Improved PointNet++
China is the largest producer and consumer of rice, and the classification of filled/unfilled rice grains is of great significance for rice breeding and genetic analysis. The traditional method for filled/unfilled rice grain identification was generally manual, which had the disadvantages of low eff...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/14/6331 |
| _version_ | 1797587516256157696 |
|---|---|
| author | Shihao Huang Zhihao Lu Yuxuan Shi Jiale Dong Lin Hu Wanneng Yang Chenglong Huang |
| author_facet | Shihao Huang Zhihao Lu Yuxuan Shi Jiale Dong Lin Hu Wanneng Yang Chenglong Huang |
| author_sort | Shihao Huang |
| collection | DOAJ |
| description | China is the largest producer and consumer of rice, and the classification of filled/unfilled rice grains is of great significance for rice breeding and genetic analysis. The traditional method for filled/unfilled rice grain identification was generally manual, which had the disadvantages of low efficiency, poor repeatability, and low precision. In this study, we have proposed a novel method for filled/unfilled grain classification based on structured light imaging and Improved PointNet++. Firstly, the 3D point cloud data of rice grains were obtained by structured light imaging. And then the specified processing algorithms were developed for the single grain segmentation, and data enhancement with normal vector. Finally, the PointNet++ network was improved by adding an additional Set Abstraction layer and combining the maximum pooling of normal vectors to realize filled/unfilled rice grain point cloud classification. To verify the model performance, the Improved PointNet++ was compared with six machine learning methods, PointNet and PointConv. The results showed that the optimal machine learning model is XGboost, with a classification accuracy of 91.99%, while the classification accuracy of Improved PointNet++ was 98.50% outperforming the PointNet 93.75% and PointConv 92.25%. In conclusion, this study has demonstrated a novel and effective method for filled/unfilled grain recognition. |
| first_indexed | 2024-03-11T00:41:00Z |
| format | Article |
| id | doaj.art-6151278a194c473aa04f51fb3020e3ae |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T00:41:00Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-6151278a194c473aa04f51fb3020e3ae2023-11-18T21:16:14ZengMDPI AGSensors1424-82202023-07-012314633110.3390/s23146331A Novel Method for Filled/Unfilled Grain Classification Based on Structured Light Imaging and Improved PointNet++Shihao Huang0Zhihao Lu1Yuxuan Shi2Jiale Dong3Lin Hu4Wanneng Yang5Chenglong Huang6College of Engineering, Huazhong Agricultural University, Wuhan 430070, ChinaCollege of Engineering, Huazhong Agricultural University, Wuhan 430070, ChinaCollege of Engineering, Huazhong Agricultural University, Wuhan 430070, ChinaCollege of Engineering, Huazhong Agricultural University, Wuhan 430070, ChinaCollege of Engineering, Huazhong Agricultural University, Wuhan 430070, ChinaNational Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan), Huazhong Agricultural University, Wuhan 430070, ChinaCollege of Engineering, Huazhong Agricultural University, Wuhan 430070, ChinaChina is the largest producer and consumer of rice, and the classification of filled/unfilled rice grains is of great significance for rice breeding and genetic analysis. The traditional method for filled/unfilled rice grain identification was generally manual, which had the disadvantages of low efficiency, poor repeatability, and low precision. In this study, we have proposed a novel method for filled/unfilled grain classification based on structured light imaging and Improved PointNet++. Firstly, the 3D point cloud data of rice grains were obtained by structured light imaging. And then the specified processing algorithms were developed for the single grain segmentation, and data enhancement with normal vector. Finally, the PointNet++ network was improved by adding an additional Set Abstraction layer and combining the maximum pooling of normal vectors to realize filled/unfilled rice grain point cloud classification. To verify the model performance, the Improved PointNet++ was compared with six machine learning methods, PointNet and PointConv. The results showed that the optimal machine learning model is XGboost, with a classification accuracy of 91.99%, while the classification accuracy of Improved PointNet++ was 98.50% outperforming the PointNet 93.75% and PointConv 92.25%. In conclusion, this study has demonstrated a novel and effective method for filled/unfilled grain recognition.https://www.mdpi.com/1424-8220/23/14/63313D structured lightpoint cloud segmentationdata enhancementdeep learningnormal vectorgrain classification |
| spellingShingle | Shihao Huang Zhihao Lu Yuxuan Shi Jiale Dong Lin Hu Wanneng Yang Chenglong Huang A Novel Method for Filled/Unfilled Grain Classification Based on Structured Light Imaging and Improved PointNet++ Sensors 3D structured light point cloud segmentation data enhancement deep learning normal vector grain classification |
| title | A Novel Method for Filled/Unfilled Grain Classification Based on Structured Light Imaging and Improved PointNet++ |
| title_full | A Novel Method for Filled/Unfilled Grain Classification Based on Structured Light Imaging and Improved PointNet++ |
| title_fullStr | A Novel Method for Filled/Unfilled Grain Classification Based on Structured Light Imaging and Improved PointNet++ |
| title_full_unstemmed | A Novel Method for Filled/Unfilled Grain Classification Based on Structured Light Imaging and Improved PointNet++ |
| title_short | A Novel Method for Filled/Unfilled Grain Classification Based on Structured Light Imaging and Improved PointNet++ |
| title_sort | novel method for filled unfilled grain classification based on structured light imaging and improved pointnet |
| topic | 3D structured light point cloud segmentation data enhancement deep learning normal vector grain classification |
| url | https://www.mdpi.com/1424-8220/23/14/6331 |
| work_keys_str_mv | AT shihaohuang anovelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT zhihaolu anovelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT yuxuanshi anovelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT jialedong anovelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT linhu anovelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT wannengyang anovelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT chenglonghuang anovelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT shihaohuang novelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT zhihaolu novelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT yuxuanshi novelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT jialedong novelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT linhu novelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT wannengyang novelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet AT chenglonghuang novelmethodforfilledunfilledgrainclassificationbasedonstructuredlightimagingandimprovedpointnet |