HarDNet: A Low Memory Traffic Network

Chao, Ping; Kao, Chao-Yang; Ruan, Yu-Shan; Huang, Ching Jer; Lin, Youn-Long

doi:10.48550/arxiv.1909.00948

Cited by 3 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we compared our proposed model with the existing related gastrointestinal polyp segmentation methods, including EnhancedUNet [ 65 ], PraNet [ 19 ], SANet [ 56 ], HarDNet [ 66 ], ACSNet [ 67 ], DBHNet [ 20 ] and CFANet [ 22 ]. We also compare BFE-Net against state-of-the-art medical image segmentation methods: U-Net [ 16 ], U-Net++ [ 17 ], DeepLabV3 [ 68 ], ResUNet [ 18 ], AttUNet [ 69 ].…”

Section: Resultsmentioning

confidence: 99%

BFE-Net: bilateral fusion enhanced network for gastrointestinal polyp segmentation

Zhang,

Hu,

et al. 2024

Biomed. Opt. Express

View full text Add to dashboard Cite

Accurate segmentation of polyp regions in gastrointestinal endoscopic images is pivotal for diagnosis and treatment. Despite advancements, challenges persist, like accurately segmenting small polyps and maintaining accuracy when polyps resemble surrounding tissues. Recent studies show the effectiveness of the pyramid vision transformer (PVT) in capturing global context, yet it may lack detailed information. Conversely, U-Net excels in semantic extraction. Hence, we propose the bilateral fusion enhanced network (BFE-Net) to address these challenges. Our model integrates U-Net and PVT features via a deep feature enhancement fusion module (FEF) and attention decoder module (AD). Experimental results demonstrate significant improvements, validating our model's effectiveness across various datasets and modalities, promising advancements in gastrointestinal polyp diagnosis and treatment.

show abstract

Section: Resultsmentioning

confidence: 99%

BFE-Net: bilateral fusion enhanced network for gastrointestinal polyp segmentation

Zhang,

Hu,

et al. 2024

Biomed. Opt. Express

View full text Add to dashboard Cite

show abstract

“…The networks in this paper are implemented on the Python-based deep learning framework PyTorch 1.11, using NVIDIA A40 with 48 G. To ensure fairness in method comparison, all network model hyperparameters are set uniformly: the number of training epochs is 100, the learning rate is 0.01, the batch size is 16, and the loss function uses crossentropy loss. The model comparisons include BiSeNetv2 [43], ConvNeXt [44], Danet [45], DDRNet [46], Deeplabv3plus [47], Unet [48], FCHarDNet [49], GCN [50], NestedUnet [51], PSPNet [52], SFNet [53], SegFormer [54], and Swin Transformer [55]. These state-of-the-art models were proved to perform well in traditional semantic segmentation-related downstream tasks.…”

Section: Implementation Detailsmentioning

confidence: 99%

A Practical Deep Learning Architecture for Large-Area Solid Wastes Monitoring Based on UAV Imagery

Liu,

Zhao,

Zhang

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

The development of global urbanization has brought about a significant amount of solid waste. These untreated wastes may be dumped in any corner, causing serious pollution to the environment. Thus, it is necessary to accurately obtain their distribution locations and detailed edge information. In this study, a practical deep learning network for recognizing solid waste piles over extensive areas using unmanned aerial vehicle (UAV) imagery has been proposed and verified. Firstly, a high-resolution dataset serving to solid waste detection was created based on UAV aerial data. Then, a dual-branch solid waste semantic segmentation model was constructed to address the characteristics of the integration of solid waste distribution with the environment and the irregular edge morphology. The Context feature branch is responsible for extracting high-level semantic features, while the Spatial feature branch is designed to capture fine-grained spatial details. After information fusion, the model obtained more comprehensive feature representation and segmentation ability. The effectiveness of the improvement was verified through ablation experiments and compared with 13 commonly used semantic segmentation models, demonstrating the advantages of the method in solid waste segmentation tasks, with an overall accuracy of over 94%, and a recall rate of 88.6%—much better than the best performing baselines. Finally, a spatial distribution map of solid waste over Jiaxing district, China was generated by the model inference, which assisted the environmental protection department in completing environmental management. The proposed method provides a feasible approach for the accurately monitoring of solid waste, so as to provide policy support for environmental protection.

show abstract

Automated measurement of cardiothoracic ratio based on semantic segmentation integration model using deep learning

Feng,

Huang,

et al. 2024

Med Biol Eng Comput

View full text Add to dashboard Cite

HarDNet: A Low Memory Traffic Network

Cited by 3 publications

References 22 publications

BFE-Net: bilateral fusion enhanced network for gastrointestinal polyp segmentation

BFE-Net: bilateral fusion enhanced network for gastrointestinal polyp segmentation

A Practical Deep Learning Architecture for Large-Area Solid Wastes Monitoring Based on UAV Imagery

Automated measurement of cardiothoracic ratio based on semantic segmentation integration model using deep learning

Contact Info

Product

Resources

About