Architecture of Swin Transformer Block. by Wen-Qing Huang (5258126)

“…Experimental results demonstrate that compared to the baseline model, the proposed enhanced model achieves a reduction of 2.5G Flops computational complexity and a 16% speed improvement on a private pavement disease dataset, with only a 1.2 percentage point decrease in AUC accuracy. Moreover, compared to other mainstream image classification models, this model exhibits more balanced performance on a public dataset, with improved accuracy and speed that better align with the practical requirements of pavement inspection. …”

Disease distribution map of the GZDL-BD. by Wen-Qing Huang (5258126)

“…Experimental results demonstrate that compared to the baseline model, the proposed enhanced model achieves a reduction of 2.5G Flops computational complexity and a 16% speed improvement on a private pavement disease dataset, with only a 1.2 percentage point decrease in AUC accuracy. Moreover, compared to other mainstream image classification models, this model exhibits more balanced performance on a public dataset, with improved accuracy and speed that better align with the practical requirements of pavement inspection. …”

Token merging module. by Wen-Qing Huang (5258126)

“…Experimental results demonstrate that compared to the baseline model, the proposed enhanced model achieves a reduction of 2.5G Flops computational complexity and a 16% speed improvement on a private pavement disease dataset, with only a 1.2 percentage point decrease in AUC accuracy. Moreover, compared to other mainstream image classification models, this model exhibits more balanced performance on a public dataset, with improved accuracy and speed that better align with the practical requirements of pavement inspection. …”

Comparative results of the ablation experiments. by Wen-Qing Huang (5258126)

“…Experimental results demonstrate that compared to the baseline model, the proposed enhanced model achieves a reduction of 2.5G Flops computational complexity and a 16% speed improvement on a private pavement disease dataset, with only a 1.2 percentage point decrease in AUC accuracy. Moreover, compared to other mainstream image classification models, this model exhibits more balanced performance on a public dataset, with improved accuracy and speed that better align with the practical requirements of pavement inspection. …”

Stir casting setup with furnace. by Charles Sarala Rubi (21608723)

Subjects:

Optical microscopic image of B₄C/Fly Ash reinforced hybrid AMCs. by Charles Sarala Rubi (21608723)

Subjects:

Experimental Results of SWR and COF (LM6/ B₄C/ Fly Ash). by Charles Sarala Rubi (21608723)

Subjects:

ANOVA for GRG. by Charles Sarala Rubi (21608723)

Subjects:

Effect of Sliding Speed on GRG. by Charles Sarala Rubi (21608723)

Subjects:

Impact of dispersoids on the Porosity of AMCs. by Charles Sarala Rubi (21608723)

Subjects:

Image of Pin on disc setup. by Charles Sarala Rubi (21608723)

Subjects:

Effect of Sliding Distance on GRG. by Charles Sarala Rubi (21608723)

Subjects:

Micrograph of worn pin of LM6  by Charles Sarala Rubi (21608723)

Subjects:

Input Parameters and their levels. by Charles Sarala Rubi (21608723)

Subjects:

Impact of B₄C & Fly ash on the Density of AMCs. by Charles Sarala Rubi (21608723)

Subjects:

Effect of Load on GRG. by Charles Sarala Rubi (21608723)

Subjects:

Surface structure of B₄C. by Charles Sarala Rubi (21608723)

Subjects:

Response table for GRG. by Charles Sarala Rubi (21608723)

Subjects:

Effect of Reinforcement Percentage on GRG. by Charles Sarala Rubi (21608723)

Subjects:

Surface structure of fly ash particles. by Charles Sarala Rubi (21608723)

Subjects: