GSCNN: a composition of CNN and Gibb Sampling computational strategy for predicting promoter in bacterial genomes

Sasikala, S.; Jeyalakshmi, T. Ratha

doi:10.1007/s41870-020-00565-y

Cited by 7 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The land type, land status, alkaline hydrolysis nitrogen,available phosphate,available potassium and other contents are measured. Figure 2 shows the structure of the layered grid sampling method [21], each small plot is 60ft×30ft through this method. In each grid, the plum blossom sampling method [22] is used for soil sampling, which is shown in Figure 3 .The soil samples of 5 points are mixed and put into soil sampling bags for labeling.…”

Section: Environmental Settingmentioning

confidence: 99%

Enhanced Backpropagation Neural Network Approach for High Precision Fertilization Method in Greenhouse Vegetable Cultivation

Tang,

Aridas,

Talip

et al. 2024

Preprint

View full text Add to dashboard Cite

The traditional method of detecting crop nutrients is based on the direct chemical detection method in the laboratory, which causes great damage to crops. In order to solve the above problems, an precision fertilization method for greenhouse vegetables based on IM-BPNN(improved backpropagation neural network) algorithm is designed in this study. First, soil samples from the farm in china are selected. With the laboratory treatment, available phosphorus, available potassium, and alkaline nitrogen are extracted. These data are preprocessed by the z-score(zero-mean normalization) standardization method. Then, the BPNN(backpropagation neural network) algorithm is improved by being trained and combined with the characteristics of the dual particle swarm optimization algorithm. After that, the soil sample data are divided into training and test sets, and the model is established by setting parameters, weights, and network hierarchy. Finally, the NBTY(nutrient balance target yield) ,BPNN(backpropagation neural network) and IM-BPNN algorithm are used to calculate the amount of fertilizer. Compared with the NBTY algorithm, the available potassium, available phosphate, and alkaline hydrolysis nitrogen increases 35.78%, 20.93% and 18.08% in the reasonable range and increases 52.09%, 37.34%, and 20.59% in the best range. Compared with the BPNN algorithm, the available potassium, available phosphate, and alkaline hydrolysis nitrogen increases 15.47%, 12.06% and 9.82% in the reasonable range and increases 19.85%,18.98% and 11.35% in the best range. It shows that the IM-BPNN algorithm can more accurately determine the amount of fertilizer required by vegetables and avoid over-application, which can improve fertilizer utilization efficiency, reduce production costs, and improve the economic feasibility of agriculture.

show abstract

Section: Environmental Settingmentioning

confidence: 99%

Enhanced Backpropagation Neural Network Approach for High Precision Fertilization Method in Greenhouse Vegetable Cultivation

Tang,

Aridas,

Talip

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…GFF [40] selected features from different scales. GSCNN [41] proposed to connect the information of two branches by the gate module. In this paper, we introduce our proposed Gate Attention module for better integrating the multi‐scale local feature and high‐level global feature.…”

Section: Related Workmentioning

confidence: 99%

Lite‐weight semantic segmentation with AG self‐attention

Liu

Gao

et al. 2023

IET Computer Vision

View full text Add to dashboard Cite

Due to the large computational and GPUs memory cost of semantic segmentation, some works focus on designing a lite weight model to achieve a good trade‐off between computational cost and accuracy. A common method is to combined CNN and vision transformer. However, these methods ignore the contextual information of multi receptive fields. And existing methods often fail to inject detailed information losses in the downsampling of multi‐scale feature. To fix these issues, we propose AG Self‐Attention, which is Enhanced Atrous Self‐Attention (EASA), and Gate Attention. AG Self‐Attention adds the contextual information of multi receptive fields into the global semantic feature. Specifically, the Enhanced Atrous Self‐Attention uses weight shared atrous convolution with different atrous rates to get the contextual information under the specific different receptive fields. Gate Attention introduces gating mechanism to inject detailed information into the global semantic feature and filter detailed information by producing “fusion” gate and “update” gate. In order to prove our insight. We conduct numerous experiments in common semantic segmentation datasets, consisting of ADE20 K, COCO‐stuff, PASCAL Context, Cityscapes, to show that our method achieves state‐of‐the‐art performance and achieve a good trade‐off between computational cost and accuracy.

show abstract