Torsional capacity evaluation of RC beams using an improved bird swarm algorithm optimised 2D convolutional neural network

Yu, Yang; Liang, Shiwei; Samali, Bijan; Nguyen, Thuc N.; Zhai, Chenxi; Li, Jianchun; Xie, Xingyang

doi:10.1016/j.engstruct.2022.115066

Cited by 64 publications

(44 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep neural networks (DNN) have shown superior performance when they are widely applied to multiple fields [ 17 , 18 , 19 , 20 , 21 , 22 ]. Figure 1 shows the architecture of our DNN.…”

Section: Methodsmentioning

confidence: 99%

Leaf-Counting in Monocot Plants Using Deep Regression Models

Xie

Walia

et al. 2023

Sensors

View full text Add to dashboard Cite

Leaf numbers are vital in estimating the yield of crops. Traditional manual leaf-counting is tedious, costly, and an enormous job. Recent convolutional neural network-based approaches achieve promising results for rosette plants. However, there is a lack of effective solutions to tackle leaf counting for monocot plants, such as sorghum and maize. The existing approaches often require substantial training datasets and annotations, thus incurring significant overheads for labeling. Moreover, these approaches can easily fail when leaf structures are occluded in images. To address these issues, we present a new deep neural network-based method that does not require any effort to label leaf structures explicitly and achieves superior performance even with severe leaf occlusions in images. Our method extracts leaf skeletons to gain more topological information and applies augmentation to enhance structural variety in the original images. Then, we feed the combination of original images, derived skeletons, and augmentations into a regression model, transferred from Inception-Resnet-V2, for leaf-counting. We find that leaf tips are important in our regression model through an input modification method and a Grad-CAM method. The superiority of the proposed method is validated via comparison with the existing approaches conducted on a similar dataset. The results show that our method does not only improve the accuracy of leaf-counting, with overlaps and occlusions, but also lower the training cost, with fewer annotations compared to the previous state-of-the-art approaches.The robustness of the proposed method against the noise effect is also verified by removing the environmental noises during the image preprocessing and reducing the effect of the noises introduced by skeletonization, with satisfactory outcomes.

show abstract

“…Deep neural networks (DNN) have shown superior performance when they are widely applied to multiple fields [ 17 , 18 , 19 , 20 , 21 , 22 ]. Figure 1 shows the architecture of our DNN.…”

Section: Methodsmentioning

confidence: 99%

Leaf-Counting in Monocot Plants Using Deep Regression Models

Xie

Walia

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…The features of lip print are rich in information, including linear, curvilinear, bifurcated, reticular, and irregular texture features, and lip print concealment is good; not easy to be copied and imitated, with uniqueness, permanence, and stability of the characteristics; an important biological feature of human identity. As a new biometric technology, lip print recognition has many advantages compared with other biometric technologies, such as high recognition rate, short recognition time, high user acceptance, and noncontact acquisition [2].…”

Section: Introductionmentioning

confidence: 99%

Lip Print Recognition Algorithm Based on Convolutional Network

Zhou

2023

Journal of Applied Mathematics

View full text Add to dashboard Cite

Identity information security is faced with various challenges, and the traditional identification technology cannot meet the needs of public security. Therefore, it is necessary to further explore and study new identification technologies. In order to solve the complex image preprocessing problems, difficult feature extraction by artificial design algorithm, and low accuracy of lip print recognition, a method based on the convolutional neural network is proposed, by building a convolutional neural network called LPRNet (Lip Print Recognition Network). The obtained lip print image is inputted into the training recognition model of the network to simplify the lip print image preprocessing. By extracting feature information and sampling operation, the model training parameters are reduced, which overcomes the difficulty of designing a complex algorithm to extract features. By analyzing and comparing the experimental results, a higher recognition rate is obtained, and the validity of the method is verified.

show abstract

“…Using ABAQUS software, a fnite element model was established to forecast the behavior of the tested beams to a level that accurately predicted the shear capabilities of the beams and recorded their failure mechanisms [9,10]. Also, using basalt microfbers in the fber-reinforced polymer is a disputed topic [11], and using optimization algorithms based on ANN to determine the torsional strength of a reinforced concrete beam [12,13].…”

Section: Introductionmentioning

confidence: 99%

Firefly Algorithm-Based Artificial Neural Network to Predict the Shear Strength in FRP-Reinforced Concrete Beams

Nikoo

Aminnejad

Lork

2023

Advances in Civil Engineering

View full text Add to dashboard Cite

The shear strength of fiber-reinforced polymer (FRP) reinforced concrete beams is often given a large safety margin by current construction requirements. Six characteristics are utilized as inputs to compute the shear strength of FRP-reinforced concrete beams. This study uses 198 samples from the literature to predict the shear strength of 139 training samples and 59 testing samples. Additionally, the ANN structure is optimized with the firefly algorithm. The FA-ANN model is also compared to ACI-440, CSA-S806, and BISE-99 codes, and the optimized model by Nehdi et al. Findings show that regarding the shear strength of FRP-reinforced concrete beams, the firefly algorithm-optimized model performs better than the other four models. Concerning accuracy, the coefficient of correlation, R2, was calculated as 0.961, while the average absolute error (AAE) is 0.22 for the shear strength of FRP-reinforced beams.

show abstract

Torsional capacity evaluation of RC beams using an improved bird swarm algorithm optimised 2D convolutional neural network

Cited by 64 publications

References 43 publications

Leaf-Counting in Monocot Plants Using Deep Regression Models

Leaf-Counting in Monocot Plants Using Deep Regression Models

Lip Print Recognition Algorithm Based on Convolutional Network

Firefly Algorithm-Based Artificial Neural Network to Predict the Shear Strength in FRP-Reinforced Concrete Beams

Contact Info

Product

Resources

About