A Diffusion-Based Detection Model for Accurate Soybean Disease Identification in Smart Agricultural Environments
Accurate detection of soybean diseases is a critical component in achieving intelligent agricultural management. However, traditional methods often underperform in complex field scenarios. This paper proposes a diffusion-based object detection model that integrates the endogenous diffusion sub-netwo...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-02-01
|
| Series: | Plants |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2223-7747/14/5/675 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850052487943290880 |
|---|---|
| author | Jiaxin Yin Weixia Li Junhong Shen Chaoyu Zhou Siqi Li Jingchao Suo Jujing Yang Ruiqi Jia Chunli Lv |
| author_facet | Jiaxin Yin Weixia Li Junhong Shen Chaoyu Zhou Siqi Li Jingchao Suo Jujing Yang Ruiqi Jia Chunli Lv |
| author_sort | Jiaxin Yin |
| collection | DOAJ |
| description | Accurate detection of soybean diseases is a critical component in achieving intelligent agricultural management. However, traditional methods often underperform in complex field scenarios. This paper proposes a diffusion-based object detection model that integrates the endogenous diffusion sub-network and the endogenous diffusion loss function to progressively optimize feature distributions, significantly enhancing detection performance for complex backgrounds and diverse disease regions. Experimental results demonstrate that the proposed method outperforms multiple baseline models, achieving a precision of 94%, recall of 90%, accuracy of 92%, and mAP@50 and mAP@75 of 92% and 91%, respectively, surpassing RetinaNet, DETR, YOLOv10, and DETR v2. In fine-grained disease detection, the model performs best on rust detection, with a precision of 96% and a recall of 93%. For more complex diseases such as bacterial blight and Fusarium head blight, precision and mAP exceed 90%. Compared to self-attention and CBAM, the proposed endogenous diffusion attention mechanism further improves feature extraction accuracy and robustness. This method demonstrates significant advantages in both theoretical innovation and practical application, providing critical technological support for intelligent soybean disease detection. |
| format | Article |
| id | doaj-art-7ade05d8cf5f4b2fa0384ea83071aaa6 |
| institution | DOAJ |
| issn | 2223-7747 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Plants |
| spelling | doaj-art-7ade05d8cf5f4b2fa0384ea83071aaa62025-08-20T02:52:48ZengMDPI AGPlants2223-77472025-02-0114567510.3390/plants14050675A Diffusion-Based Detection Model for Accurate Soybean Disease Identification in Smart Agricultural EnvironmentsJiaxin Yin0Weixia Li1Junhong Shen2Chaoyu Zhou3Siqi Li4Jingchao Suo5Jujing Yang6Ruiqi Jia7Chunli Lv8China Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaAccurate detection of soybean diseases is a critical component in achieving intelligent agricultural management. However, traditional methods often underperform in complex field scenarios. This paper proposes a diffusion-based object detection model that integrates the endogenous diffusion sub-network and the endogenous diffusion loss function to progressively optimize feature distributions, significantly enhancing detection performance for complex backgrounds and diverse disease regions. Experimental results demonstrate that the proposed method outperforms multiple baseline models, achieving a precision of 94%, recall of 90%, accuracy of 92%, and mAP@50 and mAP@75 of 92% and 91%, respectively, surpassing RetinaNet, DETR, YOLOv10, and DETR v2. In fine-grained disease detection, the model performs best on rust detection, with a precision of 96% and a recall of 93%. For more complex diseases such as bacterial blight and Fusarium head blight, precision and mAP exceed 90%. Compared to self-attention and CBAM, the proposed endogenous diffusion attention mechanism further improves feature extraction accuracy and robustness. This method demonstrates significant advantages in both theoretical innovation and practical application, providing critical technological support for intelligent soybean disease detection.https://www.mdpi.com/2223-7747/14/5/675soybean disease detectionendogenous diffusion sub-networkmulti-task optimizationdeep learningprecision agriculture |
| spellingShingle | Jiaxin Yin Weixia Li Junhong Shen Chaoyu Zhou Siqi Li Jingchao Suo Jujing Yang Ruiqi Jia Chunli Lv A Diffusion-Based Detection Model for Accurate Soybean Disease Identification in Smart Agricultural Environments Plants soybean disease detection endogenous diffusion sub-network multi-task optimization deep learning precision agriculture |
| title | A Diffusion-Based Detection Model for Accurate Soybean Disease Identification in Smart Agricultural Environments |
| title_full | A Diffusion-Based Detection Model for Accurate Soybean Disease Identification in Smart Agricultural Environments |
| title_fullStr | A Diffusion-Based Detection Model for Accurate Soybean Disease Identification in Smart Agricultural Environments |
| title_full_unstemmed | A Diffusion-Based Detection Model for Accurate Soybean Disease Identification in Smart Agricultural Environments |
| title_short | A Diffusion-Based Detection Model for Accurate Soybean Disease Identification in Smart Agricultural Environments |
| title_sort | diffusion based detection model for accurate soybean disease identification in smart agricultural environments |
| topic | soybean disease detection endogenous diffusion sub-network multi-task optimization deep learning precision agriculture |
| url | https://www.mdpi.com/2223-7747/14/5/675 |
| work_keys_str_mv | AT jiaxinyin adiffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT weixiali adiffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT junhongshen adiffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT chaoyuzhou adiffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT siqili adiffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT jingchaosuo adiffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT jujingyang adiffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT ruiqijia adiffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT chunlilv adiffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT jiaxinyin diffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT weixiali diffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT junhongshen diffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT chaoyuzhou diffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT siqili diffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT jingchaosuo diffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT jujingyang diffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT ruiqijia diffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments AT chunlilv diffusionbaseddetectionmodelforaccuratesoybeandiseaseidentificationinsmartagriculturalenvironments |