Application of deep learning-based image recognition technology to asphalt–aggregate mixtures: Methodology

Dan, Han-Cheng; Bai, Ge-Wen; Zhu, Zhihui

doi:10.1016/j.conbuildmat.2021.123770

Cited by 22 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the constructed neural network system can simulate the analysis properties of the human neural network system to a great extent. Therefore, the style exhibited by ANN when dealing with tasks is very human-friendly [ 19 ]. The basic structure of neural network technology includes input, hidden, and output layers.…”

Section: Research Theories and Methodsmentioning

confidence: 99%

Feature Recognition and Style Transfer of Painting Image Using Lightweight Deep Learning

Tan

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

This work aims to improve the feature recognition efficiency of painting images, optimize the style transfer effect of painting images, and save the cost of computer work. First, the theoretical knowledge of painting image recognition and painting style transfer is discussed. Then, lightweight deep learning techniques and their application principles are introduced. Finally, faster convolutional neural network (Faster-CNN) image feature recognition and style transfer models are designed based on a lightweight deep learning model. The model performance is comprehensively evaluated. The research results show that the designed Faster-CNN model has the highest average recognition efficiency of about 28 ms and the lowest of 17.5 ms in terms of feature recognition of painting images. The accuracy of the Faster-CNN model for image feature recognition is about 97% at the highest and 95% at the lowest. Finally, the designed Faster-CNN model can perform style recognition transfer on a variety of painting images. In terms of style recognition transfer efficiency, the highest recognition transfer rate of the designed Faster-CNN model is about 79%, and the lowest is about 77%. This work not only provides an important technical reference for feature recognition and style transfer of painting images but also contributes to the development of lightweight deep learning techniques.

show abstract

Section: Research Theories and Methodsmentioning

confidence: 99%

Feature Recognition and Style Transfer of Painting Image Using Lightweight Deep Learning

Tan

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…2 that training samples can be input in combination with the above moving image recognition process, and the input value of hidden layer neurons 26 can be calculated. At this time, the optimal model for Taekwondo moving image recognition of wearable sensors of the Internet of Things is built based on the hybrid neural network algorithm as shown in ( 6 ) below 27 . …”

Section: Design Of Wearable Sensor Of Internet Of Things Based On Hyb...mentioning

confidence: 99%

Taekwondo motion image recognition model based on hybrid neural network algorithm for wearable sensor of Internet of Things

2023

Sci Rep

View full text Add to dashboard Cite

Conventional IoT wearable sensor Taekwondo motion image recognition model mainly uses Anchor fixed proportion whole body target anchor frame to extract recognition features, which is vulnerable to dynamic noise, resulting in low displacement recognition rate of motion image. Therefore, a new IoT wearable sensor Taekwondo motion image recognition model needs to be designed based on hybrid neural network algorithm. That is, the wearable sensor Taekwondo motion image features are extracted, and the hybrid neural network algorithm is used to generate the optimization model of the wearable sensor Taekwondo motion image recognition of the Internet of Things, so as to achieve effective recognition of Taekwondo motion images. The experimental results show that the designed wearable sensor of the Internet of Things based on the hybrid neural network algorithm has a high recognition rate of the motion image displacement of the Taekwondo motion image recognition model, which proves that the designed Taekwondo motion image recognition model has good recognition effect, reliability, and certain application value, and has made certain contributions to optimizing the Taekwondo movement.

show abstract

“…Representative models include VGGNet 15 , ResNet 16 , MobileNet 17 , ShuffleNet 18 , and Efficient 19 . Dan 20 uses a convolutional neural network (U-NET++) to segment aggregated particles, which can accurately obtain the geometry of the particles appearing on the side of the asphalt mixture. An identification and analysis method of asphalt mixture segregation and weakening is proposed.…”

Section: Introductionmentioning

confidence: 99%

ConvNeXt steel slag sand substitution rate detection method incorporating attention mechanism

Teng

Zhu

et al. 2023

Sci Rep

View full text Add to dashboard Cite

The proportion of natural sand replaced by steel slag sand affects the volumetric stability of steel slag mortar and steel slag concrete. However, the steel slag substitution rate detection method is inefficient and lacks representative sampling. Therefore, a deep learning-based steel slag sand substitution rate detection method is proposed. The technique adds a squeeze and excitation (SE) attention mechanism to the ConvNeXt model to improve the model's efficiency in extracting the color features of steel slag sand mix. Meanwhile, the model's accuracy is further enhanced by using the migration learning method. The experimental results show that SE can effectively help ConvNeXt acquire images' color features. The model's accuracy in predicting the replacement rate of steel slag sand is 87.99%, which is better than the original ConvNeXt network and other standard convolutional neural networks. After using the migration learning training method, the model predicts the steel slag sand substitution rate with 92.64% accuracy, improving accuracy by 4.65%. The SE attention mechanism and the migration learning training method can help the model acquire the critical features of the image better and effectively improve the model's accuracy. The method proposed in this paper can identify the steel slag sand substitution rate quickly and accurately and can be used for the detection of the steel slag sand substitution rate.

show abstract

Application of deep learning-based image recognition technology to asphalt–aggregate mixtures: Methodology

Cited by 22 publications

References 27 publications

Feature Recognition and Style Transfer of Painting Image Using Lightweight Deep Learning

Feature Recognition and Style Transfer of Painting Image Using Lightweight Deep Learning

Taekwondo motion image recognition model based on hybrid neural network algorithm for wearable sensor of Internet of Things

ConvNeXt steel slag sand substitution rate detection method incorporating attention mechanism

Contact Info

Product

Resources

About