Ripe-Detection: A Lightweight Method for Strawberry Ripeness Detection
Strawberry (<i>Fragaria</i> × <i>ananassa</i>), a nutrient-dense fruit with significant economic value in commercial cultivation, faces critical detection challenges in automated harvesting due to complex growth conditions such as foliage occlusion and variable illumination....
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-07-01
|
| Series: | Agronomy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4395/15/7/1645 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849304090386890752 |
|---|---|
| author | Helong Yu Cheng Qian Zhenyang Chen Jing Chen Yuxin Zhao |
| author_facet | Helong Yu Cheng Qian Zhenyang Chen Jing Chen Yuxin Zhao |
| author_sort | Helong Yu |
| collection | DOAJ |
| description | Strawberry (<i>Fragaria</i> × <i>ananassa</i>), a nutrient-dense fruit with significant economic value in commercial cultivation, faces critical detection challenges in automated harvesting due to complex growth conditions such as foliage occlusion and variable illumination. To address these limitations, this study proposes Ripe-Detection, a novel lightweight object detection framework integrating three key innovations: a PEDblock detection head architecture with depth-adaptive feature learning capability, an ADown downsampling method for enhanced detail perception with reduced computational overhead, and BiFPN-based hierarchical feature fusion with learnable weighting mechanisms. Developed using a purpose-built dataset of 1021 annotated strawberry images (<i>Fragaria</i> × <i>ananassa</i> ‘Red Face’ and ‘Sachinoka’ varieties) from Changchun Xiaohongmao Plantation and augmented through targeted strategies to enhance model robustness, the framework demonstrates superior performance over existing lightweight detectors, achieving mAP50 improvements of 13.0%, 9.2%, and 3.9% against YOLOv7-tiny, YOLOv10n, and YOLOv11n, respectively. Remarkably, the architecture attains 96.4% mAP50 with only 1.3M parameters (57% reduction from baseline) and 4.4 GFLOPs (46% lower computation), simultaneously enhancing accuracy while significantly reducing resource requirements, thereby providing a robust technical foundation for automated ripeness assessment and precision harvesting in agricultural robotics. |
| format | Article |
| id | doaj-art-1d465094dc004239b34d217d77cf67e2 |
| institution | Kabale University |
| issn | 2073-4395 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj-art-1d465094dc004239b34d217d77cf67e22025-08-20T03:55:49ZengMDPI AGAgronomy2073-43952025-07-01157164510.3390/agronomy15071645Ripe-Detection: A Lightweight Method for Strawberry Ripeness DetectionHelong Yu0Cheng Qian1Zhenyang Chen2Jing Chen3Yuxin Zhao4College of Information Technology, Jilin Agricultural University, Changchun 130118, ChinaCollege of Information Technology, Jilin Agricultural University, Changchun 130118, ChinaSmart Agriculture Research Institute, Jilin Agricultural University, Changchun 130118, ChinaCollege of Information Technology, Jilin Agricultural University, Changchun 130118, ChinaSchool of Municipal and Environmental Engineering, Jilin Jianzhu University, Changchun 130000, ChinaStrawberry (<i>Fragaria</i> × <i>ananassa</i>), a nutrient-dense fruit with significant economic value in commercial cultivation, faces critical detection challenges in automated harvesting due to complex growth conditions such as foliage occlusion and variable illumination. To address these limitations, this study proposes Ripe-Detection, a novel lightweight object detection framework integrating three key innovations: a PEDblock detection head architecture with depth-adaptive feature learning capability, an ADown downsampling method for enhanced detail perception with reduced computational overhead, and BiFPN-based hierarchical feature fusion with learnable weighting mechanisms. Developed using a purpose-built dataset of 1021 annotated strawberry images (<i>Fragaria</i> × <i>ananassa</i> ‘Red Face’ and ‘Sachinoka’ varieties) from Changchun Xiaohongmao Plantation and augmented through targeted strategies to enhance model robustness, the framework demonstrates superior performance over existing lightweight detectors, achieving mAP50 improvements of 13.0%, 9.2%, and 3.9% against YOLOv7-tiny, YOLOv10n, and YOLOv11n, respectively. Remarkably, the architecture attains 96.4% mAP50 with only 1.3M parameters (57% reduction from baseline) and 4.4 GFLOPs (46% lower computation), simultaneously enhancing accuracy while significantly reducing resource requirements, thereby providing a robust technical foundation for automated ripeness assessment and precision harvesting in agricultural robotics.https://www.mdpi.com/2073-4395/15/7/1645computer visiondeep learningimage processingprecision agriculturestrawberry ripeness detection |
| spellingShingle | Helong Yu Cheng Qian Zhenyang Chen Jing Chen Yuxin Zhao Ripe-Detection: A Lightweight Method for Strawberry Ripeness Detection Agronomy computer vision deep learning image processing precision agriculture strawberry ripeness detection |
| title | Ripe-Detection: A Lightweight Method for Strawberry Ripeness Detection |
| title_full | Ripe-Detection: A Lightweight Method for Strawberry Ripeness Detection |
| title_fullStr | Ripe-Detection: A Lightweight Method for Strawberry Ripeness Detection |
| title_full_unstemmed | Ripe-Detection: A Lightweight Method for Strawberry Ripeness Detection |
| title_short | Ripe-Detection: A Lightweight Method for Strawberry Ripeness Detection |
| title_sort | ripe detection a lightweight method for strawberry ripeness detection |
| topic | computer vision deep learning image processing precision agriculture strawberry ripeness detection |
| url | https://www.mdpi.com/2073-4395/15/7/1645 |
| work_keys_str_mv | AT helongyu ripedetectionalightweightmethodforstrawberryripenessdetection AT chengqian ripedetectionalightweightmethodforstrawberryripenessdetection AT zhenyangchen ripedetectionalightweightmethodforstrawberryripenessdetection AT jingchen ripedetectionalightweightmethodforstrawberryripenessdetection AT yuxinzhao ripedetectionalightweightmethodforstrawberryripenessdetection |