相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Forestry
Jingwen Huang et al.
Summary: In order to reduce forest fires, we propose an improved early forest fire smoke detection model based on deformable transformer, which includes the best sparse spatial sampling for smoke and the transformer's relation modeling capability. We also introduce modules for perceiving features of small or inconspicuous smoke, and propose an iterative bounding box combination method for accurate bounding box generation. Experimental results show that our model significantly improves the accuracy of forest fire smoke detection compared to mainstream models.
Article
Chemistry, Multidisciplinary
Hui Luo et al.
Summary: Automatic pavement crack detection is crucial for reducing road maintenance costs and ensuring transportation safety. A new architecture named STrans-YOLOX is proposed to address the challenges of modeling long-range dependencies and irregular crack shapes. The proposed method achieves state-of-the-art performance on a challenging pavement crack dataset.
APPLIED SCIENCES-BASEL
(2023)
Article
Environmental Sciences
Xiangzeng Liu et al.
Summary: In this paper, a novel model named RoadFormer is proposed to accurately detect and extract roads using remote sensing technology. The model adopts a Swin Transformer as the backbone to effectively extract long-range information, and incorporates innovative bottleneck modules and a lightweight decoder to enhance feature representation and generate extraction results. Experimental results demonstrate the advantages of RoadFormer over comparable methods on the Deepglobe and Massachusetts datasets.
Article
Chemistry, Multidisciplinary
Minmin Yu et al.
Summary: Transformer models have achieved great results in computer vision, but there are few studies on their performance in remote sensing. This paper compares three mainstream transformer models and evaluates their performance in remote-sensing segmentation. The experimental results show that different transformer models have varying segmentation performance on different scales of remote-sensing data sets.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Applied
Kaizhi Zheng et al.
Summary: The study found that mutton from intensively fattened sheep was more tender than that from traditionally raised sheep. Proteomic analysis revealed 49 differentially expressed proteins in the longissimus dorsi muscle, and after bioinformatics analysis, 5 cytoskeletal proteins, 3 protein binding proteins, and 7 metabolic enzymes were identified as potential biomarkers for mutton tenderness. The expression of these proteins was verified using parallel reaction monitoring (PRM). Cytoskeletal proteins and metabolic enzymes play a central role in determining mutton tenderness.
Article
Agronomy
Yue Gu et al.
Summary: This study used machine vision technology for automatic recognition of important behaviors of cage-reared ducks, and found that the neck extension, trampling, and wing spread behaviors of laying ducks significantly increase when they feel panic and fear.
Article
Computer Science, Software Engineering
Wenhai Wang et al.
Summary: This work presents the improved Pyramid Vision Transformer v2 (PVT v2) by adding three designs, achieving significant improvements in fundamental vision tasks. PVT v2 performs comparably or better than recent work such as the Swin transformer.
COMPUTATIONAL VISUAL MEDIA
(2022)
Article
Chemistry, Analytical
Huantong Geng et al.
Summary: This paper proposes an unsupervised domain adaptive object detection method called CA-DINO, which addresses the challenges of lacking annotated data and varying imaging conditions in new scenarios. The method utilizes attention-enhanced double discriminators and weak restraints on category-level tokens to aggregate and align context information in feature representations, reducing domain discrepancy. Experimental results demonstrate the effectiveness of this approach.
Article
Chemistry, Applied
Cunchuan Liu et al.
Summary: The combination of electronic nose and hyperspectral image was used to evaluate mutton freshness. The proposed method provides an approach for accurately analyzing mutton freshness and can be used as a technical basis for investigating other meat qualities.
Article
Agronomy
Dashuai Wang et al.
Summary: In precision agriculture, unmanned aerial vehicles (UAVs) play a crucial role in detecting obstacles in farmland environments. Object detection algorithms based on convolutional neural networks (CNN) and transformer architectures, such as Detection Transformer (DETR) and Deformable DETR, have achieved remarkable results. This study introduces the Non-local Deformable DETR, an improved version that combines global modeling capability and front-end ResNet to enhance the performance of farmland obstacle detection. Comparative experiments show that Non-local Deformable DETR has a higher mAP value compared to the original Deformable DETR, demonstrating its effectiveness in detecting obstacles, particularly small and slender objects.
Article
Food Science & Technology
Binbin Fan et al.
Summary: This study proposes a method based on near-infrared hyperspectral imaging, machine learning, and sparrow search algorithm for classification and quantification of adulterated mutton under the effect of mutton flavor essence. The performance of the models is improved through optimization, achieving high accuracy in the classification and quantification of mutton adulteration.
Article
Computer Science, Interdisciplinary Applications
Gabriel Iluebe Okolo et al.
Summary: This study examines the application of a novel enhanced Vision Transformer model (IEViT) for the task of chest X-ray image classification. The proposed model outperforms the state-of-the-art Vision Transformer (ViT) model on multiple datasets, demonstrating its superior performance and generalization ability for assisting diagnosis using chest X-ray images.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2022)
Article
Agriculture, Multidisciplinary
Bowen Niu et al.
Summary: This paper proposes a novel semantic segmentation model (HSI-TransUNet) for crop mapping using UAV hyperspectral imagery. The model utilizes the abundant spatial and spectral information of the HSI data and achieves good performance, with the effectiveness of each refined module validated by experiments.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Geography, Physical
Qibin He et al.
Summary: The article proposes a hierarchical graph network called GraFNet for multimodal semantic segmentation in remote sensing scenes. It addresses the challenges of object diversity and cross-modal gap by introducing a new modeling paradigm and utilizing semantic topological graphs. Extensive experiments demonstrate that GraFNet outperforms existing methods on various datasets.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2022)
Article
Food Science & Technology
Yaoxin Zhang et al.
Summary: This paper proposes a method based on recurrence plot transformed by spectrum combined with convolutional neural network (RP-CNN) to detect adulterated mutton, and predicts the classification and pork content of adulterated mutton. The experimental results show that this method has high accuracy and comparability in the classification and pork content prediction of adulterated mutton.
Article
Computer Science, Artificial Intelligence
S. Naveen et al.
Summary: This study explores the combination of Transformer models (such as BERT, GPT2, XLNet) with AttnGAN to improve the performance of text to image generation, proposing the AttnGAN(TRANS) architecture. Experimental results demonstrate a significant improvement over traditional methods, with the potential to outperform contemporary state-of-the-art approaches.
IMAGE AND VISION COMPUTING
(2021)
Article
Engineering, Chemical
Shida Zhao et al.
Summary: This article presents a real-time classification and detection method for mutton parts based on a single shot detector (SSD), using transfer learning to train an optimized model. Experimental results show that SSD-MobileNetV1 demonstrates high accuracy and good real-time performance compared to other methods.
JOURNAL OF FOOD PROCESS ENGINEERING
(2021)
Article
Chemistry, Analytical
Jun Chen et al.
Review
Agriculture, Multidisciplinary
Jiang Xinhua et al.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2018)
