Advancing Rice Grain Impurity Segmentation with an Enhanced SegFormer and Multi-Scale Feature Integration
During the rice harvesting process, severe occlusion and adhesion exist among multiple targets, such as rice, straw, and leaves, making it difficult to accurately distinguish between rice grains and impurities. To address the current challenges, a lightweight semantic segmentation algorithm for impu...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/27/1/70 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588566514892800 |
|---|---|
| author | Xiulin Qiu Hongzhi Yao Qinghua Liu Hongrui Liu Haozhi Zhang Mengdi Zhao |
| author_facet | Xiulin Qiu Hongzhi Yao Qinghua Liu Hongrui Liu Haozhi Zhang Mengdi Zhao |
| author_sort | Xiulin Qiu |
| collection | DOAJ |
| description | During the rice harvesting process, severe occlusion and adhesion exist among multiple targets, such as rice, straw, and leaves, making it difficult to accurately distinguish between rice grains and impurities. To address the current challenges, a lightweight semantic segmentation algorithm for impurities based on an improved SegFormer network is proposed. To make full use of the extracted features, the decoder was redesigned. First, the Feature Pyramid Network (FPN) was introduced to optimize the structure, selectively fusing the high-level semantic features and low-level texture features generated by the encoder. Secondly, a Part Large Kernel Attention (Part-LKA) module was designed and introduced after feature fusion to help the model focus on key regions, simplifying the model and accelerating computation. Finally, to compensate for the lack of spatial interaction capabilities, Bottleneck Recursive Gated Convolution (B-<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi mathvariant="normal">g</mi><mi>n</mi></msup></semantics></math></inline-formula>Conv) was introduced to achieve effective segmentation of rice grains and impurities. Compared with the original model, the improved model’s pixel accuracy (PA) and F1 score increased by 1.6% and 3.1%, respectively. This provides a valuable algorithmic reference for designing a real-time impurity rate monitoring system for rice combine harvesters. |
| format | Article |
| id | doaj-art-92eab50774fc4af682419dee04d7535b |
| institution | Kabale University |
| issn | 1099-4300 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj-art-92eab50774fc4af682419dee04d7535b2025-01-24T13:31:54ZengMDPI AGEntropy1099-43002025-01-012717010.3390/e27010070Advancing Rice Grain Impurity Segmentation with an Enhanced SegFormer and Multi-Scale Feature IntegrationXiulin Qiu0Hongzhi Yao1Qinghua Liu2Hongrui Liu3Haozhi Zhang4Mengdi Zhao5School of Automation, Jiangsu University of Science and Technology, Zhenjiang 212100, ChinaSchool of Computer, Jiangsu University of Science and Technology, Zhenjiang 212100, ChinaSchool of Automation, Jiangsu University of Science and Technology, Zhenjiang 212100, ChinaSchool of Computer, Jiangsu University of Science and Technology, Zhenjiang 212100, ChinaSchool of Computer, Jiangsu University of Science and Technology, Zhenjiang 212100, ChinaCollege of Materials Science and Engineering, Suzhou University of Science and Technology, Suzhou 215011, ChinaDuring the rice harvesting process, severe occlusion and adhesion exist among multiple targets, such as rice, straw, and leaves, making it difficult to accurately distinguish between rice grains and impurities. To address the current challenges, a lightweight semantic segmentation algorithm for impurities based on an improved SegFormer network is proposed. To make full use of the extracted features, the decoder was redesigned. First, the Feature Pyramid Network (FPN) was introduced to optimize the structure, selectively fusing the high-level semantic features and low-level texture features generated by the encoder. Secondly, a Part Large Kernel Attention (Part-LKA) module was designed and introduced after feature fusion to help the model focus on key regions, simplifying the model and accelerating computation. Finally, to compensate for the lack of spatial interaction capabilities, Bottleneck Recursive Gated Convolution (B-<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi mathvariant="normal">g</mi><mi>n</mi></msup></semantics></math></inline-formula>Conv) was introduced to achieve effective segmentation of rice grains and impurities. Compared with the original model, the improved model’s pixel accuracy (PA) and F1 score increased by 1.6% and 3.1%, respectively. This provides a valuable algorithmic reference for designing a real-time impurity rate monitoring system for rice combine harvesters.https://www.mdpi.com/1099-4300/27/1/70riceimpuritiessemantic segmentationSegFormer |
| spellingShingle | Xiulin Qiu Hongzhi Yao Qinghua Liu Hongrui Liu Haozhi Zhang Mengdi Zhao Advancing Rice Grain Impurity Segmentation with an Enhanced SegFormer and Multi-Scale Feature Integration Entropy rice impurities semantic segmentation SegFormer |
| title | Advancing Rice Grain Impurity Segmentation with an Enhanced SegFormer and Multi-Scale Feature Integration |
| title_full | Advancing Rice Grain Impurity Segmentation with an Enhanced SegFormer and Multi-Scale Feature Integration |
| title_fullStr | Advancing Rice Grain Impurity Segmentation with an Enhanced SegFormer and Multi-Scale Feature Integration |
| title_full_unstemmed | Advancing Rice Grain Impurity Segmentation with an Enhanced SegFormer and Multi-Scale Feature Integration |
| title_short | Advancing Rice Grain Impurity Segmentation with an Enhanced SegFormer and Multi-Scale Feature Integration |
| title_sort | advancing rice grain impurity segmentation with an enhanced segformer and multi scale feature integration |
| topic | rice impurities semantic segmentation SegFormer |
| url | https://www.mdpi.com/1099-4300/27/1/70 |
| work_keys_str_mv | AT xiulinqiu advancingricegrainimpuritysegmentationwithanenhancedsegformerandmultiscalefeatureintegration AT hongzhiyao advancingricegrainimpuritysegmentationwithanenhancedsegformerandmultiscalefeatureintegration AT qinghualiu advancingricegrainimpuritysegmentationwithanenhancedsegformerandmultiscalefeatureintegration AT hongruiliu advancingricegrainimpuritysegmentationwithanenhancedsegformerandmultiscalefeatureintegration AT haozhizhang advancingricegrainimpuritysegmentationwithanenhancedsegformerandmultiscalefeatureintegration AT mengdizhao advancingricegrainimpuritysegmentationwithanenhancedsegformerandmultiscalefeatureintegration |