Semantic segmentation based on Deep learning for the detection of Cyanobacterial Harmful Algal Blooms (CyanoHABs) using synthetic images

Barrientos-Espillco, Fredy; Gascó, Esther; López-González, Clara I.; Gómez-Silva, María José; Pajares, Gonzalo

doi:10.1016/j.asoc.2023.110315

Cited by 10 publications

(4 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This was repeated on the right path until the resolution of the feature maps matched that of the input images. 54 Once the model had been trained, it was applied in a test set that cut city-level satellite imagery across the country to obtain the national building rooftop areas and PV installation potential.…”

Section: Methodsmentioning

confidence: 99%

Achieving net zero energy heating by integrating the building envelope as a thermal battery

Zhi,

Sun,

Gao

et al. 2024

iScience

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 99%

Achieving net zero energy heating by integrating the building envelope as a thermal battery

Zhi,

Sun,

Gao

et al. 2024

iScience

View full text Add to dashboard Cite

“…Didapatkan hasil akurasi model meningkat setelah menggunakan augmentasi, yaitu dari 74% menjadi 87.15% [17]. Penelitian lain oleh Fredy Barrientos-Espillco, dkk juga membuktikan bahwa model UNet++ dengan EfficientNet-B6 dengan augmentasi data, mencapai generalisasi dan performa yang baik [18]. Lozhkin, dkk juga melakukan penelitian menggunakan model EfficientNet-B6 dan augmentasi data, dan didapatkan kesimpulan augemntasi data merupakan pendekatan yang efektif dalam memecahkan masalah segmentasi semantik [19].…”

Section: Pendahuluanunclassified

Klasifikasi Citra Daging Sapi dan Daging Babi Menggunakan CNN Arsitektur EfficientNet-B6 dan Augmentasi Data

Martias,

Jasril,

Sanjaya

et al. 2023

json

View full text Add to dashboard Cite

In daily life, beef often serves as a staple food for humans. However, the high and expensive price of beef has prompted traders to adulterate it with pork for the sake of profit. Such adulteration has serious implications in the Islamic religion, where not all types of meat are considered halal (permissible for consumption), such as pork. As a result, consumers often remain unaware that the beef they purchase has been adulterated with pork. At a glance, both types of meat exhibit similar appearance and texture, making them difficult to differentiate. This research aims to classify beef and pork using a deep learning model with the Convolutional Neural Network (CNN) method, combined with data augmentation. The model used is EfficientNet-B6 with variations in the testing scenario. The variations include the ratio of training and testing data, learning rates, and optimizer for EfficientNet-B6. Data augmentation is performed using techniques such as random rotation, shifting, image scaling, vertical and horizontal flipping, and nearest pixel filling. Evaluation results using the confusion matrix show that the model with data augmentation achieves the highest accuracy for the classes of beef, pork, and adulterated samples at 92.00%, while the model without augmentation achieves an accuracy of 91.67%. However, from this experiment, the best scenario to avoid misclassifying pork and adulterated samples as beef can be obtained. This scenario involves a model with data augmentation, a 90:10 data split, SGD optimizer, and a learning rate of 0.01, which achieves the highest precision for the beef class at 96.05%. The research findings demonstrate that the use of data augmentation on images can improve the model's performance, and the model with data augmentation, a 90:10 data split, SGD optimizer, and a learning rate of 0.01 exhibits the best performance in classifying beef images.

show abstract

“…Furthermore, the high cost and large size of multi-functional spectrometers [23][24][25] make them impractical to acquire data in research groups with limited funding. Therefore, Generative Adversarial Networks (GAN) can be used for data augmentation [26][27][28][29][30][31][32]. Cao Z et al [33] combined GAN networks in spectral data analysis to enhance analysis accuracy and mitigate overfitting.…”

Section: Introductionmentioning

confidence: 99%

Research on the chemical oxygen demand spectral inversion model in water based on IPLS-GAN-SVM hybrid algorithm

Lu,

Zou,

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

Spectral collinearity and limited spectral datasets are the problems influencing Chemical Oxygen Demand (COD) modeling. To address the first problem and obtain optimal modeling range, the spectra are preprocessed using six methods including Standard Normal Variate, Savitzky-Golay Smoothing Filtering (SG) etc. Subsequently, the 190–350 nm spectral range is divided into 10 subintervals, and Interval Partial Least Squares (IPLS) is used to perform PLS modeling on each interval. The results indicate that it is best modeled in the 7th range (238~253 nm). The values of Mean Square Error (MSE), Mean Absolute Error (MAE) and R2score of the model without pretreatment are 1.6489, 1.0661, and 0.9942. After pretreatment, the SG is better than others, with MSE and MAE decreasing to 1.4727, 1.0318 and R2score improving to 0.9944. Using the optimal model, the predicted COD for three samples are 10.87 mg/L, 14.88 mg/L, and 19.29 mg/L. To address the problem of the small dataset, using Generative Adversarial Networks for data augmentation, three datasets are obtained for Support Vector Machine (SVM) modeling. The results indicate that, compared to the original dataset, the SVM’s MSE and MAE have decreased, while its accuracy has improved by 2.88%, 11.53%, and 11.53%, and the R2score has improved by 18.07%, 17.40%, and 18.74%.

show abstract

Semantic segmentation based on Deep learning for the detection of Cyanobacterial Harmful Algal Blooms (CyanoHABs) using synthetic images

Cited by 10 publications

References 37 publications

Achieving net zero energy heating by integrating the building envelope as a thermal battery

Achieving net zero energy heating by integrating the building envelope as a thermal battery

Klasifikasi Citra Daging Sapi dan Daging Babi Menggunakan CNN Arsitektur EfficientNet-B6 dan Augmentasi Data

Research on the chemical oxygen demand spectral inversion model in water based on IPLS-GAN-SVM hybrid algorithm

Contact Info

Product

Resources

About