A novel detection and segmentation system for eichhornia crassipes growth rate using region vision transformer-based adaptive Yolo with Unet
Eichhornia crassipes has an extreme rapid growth rate, and has capability to doubling its population in approximately two to three weeks under optimal conditions, with biomass accumulation. It is required to protect the safety of water resources with rigorous ecological environment by detecting Eich...
Saved in:
| Published in: | Aquacultural engineering Vol. 112; p. 102659 |
|---|---|
| Main Authors: | , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier B.V
15.01.2026
|
| Subjects: | |
| ISSN: | 0144-8609 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Eichhornia crassipes has an extreme rapid growth rate, and has capability to doubling its population in approximately two to three weeks under optimal conditions, with biomass accumulation. It is required to protect the safety of water resources with rigorous ecological environment by detecting Eichhornia crassipes using deep learning techniques. The conventional techniques of Eichhornia crassipes paves more attention to reduce the growth rate of Eichhornia crassipes on the aquatic environment, but real-time monitoring of large areas is critical and it becomes challenging issue. Moreover, conventional models provide low accuracy and poor detection results on target species with unclear satellite image characteristics. Hence, to overcome these issues a new developed strategy for segmenting the Eichhornia crassipes is implemented to identify the growth rate and formulate effective control strategies. Initially, the required multi-spectral images are fetched from the distinct standard datasets for recognizing the Eichhornia crassipes growth rate. The acquired images are fed to the Region-Vision Transformer-based Adaptive Yolo (RViT-Yolo-Unet++) for performing joint detection and segmentation. The RViT-Yolo detection model captures long-range contextual dependencies and global relationships, which leads to more comprehensive understanding of complex patterns. RViT-Yolo improves the accuracy and speed by adapting the visual characteristics of the Eichhornia crassipes. While, the Unet++ segmentation model provide better growth rate through resource partitioning, and allows for propagation. The RViT-Yolo-Unet++ segmentation process manage its invasive growth by identifying and removing irrelevant boundaries of the image, leading improved water quality, enhanced biofuel production, and optimized pollution control. The detection and segmentation networks are serially connected to provide efficient growth rate detection results. Initially, the garnered images are applied to the Eichhornia crassipes detection module, where the RViT-Yolo is utilized for detecting the Eichhornia crassipes in a specific region. Moreover, for improving the detection performance of Eichhornia crassipes, the parameters of Yolo are optimally selected by an Improved Crayfish Optimization Algorithm (TI-COA). While optimizing Eichhornia crassipes, it enhances growth rate, and provides greater quantity of biomass production to recover the resources. Also, it improves the water quality and reduces the operating costs and making wastewater treatment more economically feasible. Subsequently, the detected region images are passed to the segmentation module, where the Unet++ model is used for segmenting the Eichhornia crassipes affected regions that help to discover the Eichhornia crassipes growth rate. Finally, the research experiments are performed for the implemented framework by comparing the best measure of accuracy shows 7.98 % Unet, 7.52 % Unet3 + , 7.06 % ResUnet, and 4.83 % Trans-Unet. Also, proposed model achieves 95.77 % superior result for accuracy using detection model. Here, the developed RViT-Yolo-Unet++ has achieved better outcome than existing models. Moreover, the proposed model ensures its superior performance in the growth rate detection of Eichhornia crassipes. Eichhornia crassipes with high growth rate quickly generates large biomass. It suggests several advantages by enabling efficient pollutant removal for water purification, providing rich biomass for biofuel production and other eco-friendly products, and creates an organic material for soil. |
|---|---|
| ISSN: | 0144-8609 |
| DOI: | 10.1016/j.aquaeng.2025.102659 |