Deep Learning Approach for the Morphological Differentiation of Corn Seed Types

Artificial intelligence: finding the intersection of predictive modeling and clinical utility

Tunnel boring machine vibration-based deep learning for the ground identification of working faces

Development of hybrid models based on deep learning and optimized machine learning algorithms for brain tumor Multi-Classification

Near real-time hurricane rainfall forecasting using convolutional neural network models with Integrated Multi-satellitE Retrievals for GPM (IMERG) product

INTRODUCTION: When Diabetic Retinopathy (DR) is not identified promptly; it frequently results in sight impairment. To properly diagnose and treat DR, preprocessing of picture methods and precise prediction models are essential. With the help of numerous well-liked filters and a Deep CNN (Convolutional Neural Network) model, the comprehensive method for DR image preparation and prognosis presented in this research is described. Using the filters that focus boundaries and contours in the ocular pictures is the first step in the initial processing stage. This procedure tries to find anomalies linked to DR. By the usage of filters, the excellence of pictures can be developed and minimize disturbances, preserving critical information. The Deep CNN algorithm has been trained to generate forecasts on the cleaned retinal pictures following the phase of preprocessing. The filters efficiently eliminate interference without sacrificing vital data. Convolutional type layers, pooling type layers, and fully associated layers are used in the CNN framework, which was created especially for image categorization tasks, to acquire data and understand the relationships associated with DR. OBJECTIVES: Using image preprocessing techniques such as the Sobel, Wiener, Gaussian, and non-local mean filters is a promising approach for DR analysis. Then, predicting using a CNN completes the approach. These preprocessing filters enhance the images and prepare them for further examination. The pre-processed images are fed into a CNN model. The model extracts significant information from the images by identifying complex patterns. DR or classification may be predicted by the CNN model through training on a labeled dataset. METHODS: The Method Preprocessing is employed for enhancing the clarity and difference of retina fundus picture by removing noise and fluctuation. The preprocessing stage is utilized for the normalization of the pictures and non-uniform brightness adjustment in addition to contrast augmentation and noise mitigation to remove noises and improve the rate of precision of the subsequent processing stages. RESULTS: To improve image quality and reduce noise, preprocessing techniques including Sobel, Wiener, Gaussian, and non-local mean filters are frequently employed in image processing jobs. For a particular task, the non-local mean filter produces superior results; for enhanced performance, it may be advantageous to combine it with a CNN. Before supplying the processed images to the CNN for prediction, the non-local mean filter can assist reduce noise and improve image details. CONCLUSION: A promising method for DR analysis entails the use of image preprocessing methods such as the Sobel, Wiener, Gaussian, and non-local mean filters, followed by prediction using a CNN. These preprocessing filters improve the photos and get them ready for analysis. After being pre-processed, the photos are sent into a CNN model, which uses its capacity to discover intricate patterns to draw out important elements from the images. The CNN model may predict DR or classification by training it on a labeled dataset. The development of computer-aided diagnosis systems for DR is facilitated by the integration of CNN prediction with image preprocessing filters. This strategy may increase the effectiveness of healthcare workers, boost patient outcomes, and lessen the burden of DR.

Corn is one of the essential commodities in agriculture. All components of corn can be utilized and accommodated for the benefit of humans. One of the supporting components is the quality of corn seeds, where a specific source has the physiological qualities to survive. The problem is how to get information on the quality of corn seeds at agricultural locations and get information through the physical image alone. This research tries to find a solution to obtain high accuracy in classifying corn kernels using a convolutional neural network because there is a profound training process. The problem with convolutional neural networks is the training process takes a long time, depending on the number of layers in the architecture. This research contributes to increasing the computing time with the proposed contribution by adding Region proposals with a convex hull to use on a custom layer. The method's purpose is a region proposal area with a convex hull to increase the focus on the convolution multiplication process. It affected reducing unnecessary objects in background images. A custom layer architecture by maintaining the priority layer is an option to get a shorter computational time in constructing a model. In addition, the architecture that is made still considers the stability of the training process. The results on the classification of corn seeds are obtained by a model with an average accuracy of 99.01%—the Computational training time to get the model is 2 minutes 30 seconds. The average error value for MSE is 0.0125, RMSE is 0.118, and MAE is 0.0108. The experimental data testing process has an accuracy ranging from 77% -99%. In conclusion, using region proposals can increase accuracy by around 0.3% because focused objects assist the convolution process

Landslide is a natural disaster that seriously affects human life and social development. In this study, the characteristics and effectiveness of convolutional neural network (CNN) and conventional machine learning (ML) methods in a landslide susceptibility assessment (LSA) are compared. Six ML methods used in this study are Adaboost, multilayer perceptron neural network (MLP-NN), random forest (RF), naive Bayes, decision tree (DT), and gradient boosting decision tree (GBDT). First, the basic knowledge and structures of the CNN and ML methods, and the steps of the LSA are introduced. Then, 11 conditioning factors in three categories in the Hongxi River Basin, Pingwu County, Mianyang City, Sichuan Province are chosen to build the train, validation, and test samples. The CNN and ML models are constructed based on these samples. For comparison, indicator methods, statistical methods, and landslide susceptibility maps (LSMs) are used. The result shows that the CNN can obtain the highest accuracy (86.41%) and the highest AUC (0.9249) in the LSA. The statistical methods represented by the mean and variance of TP and TN perform more firmly on the possibility of landslide occurrence. Furthermore, the LSMs show that all models can successfully identify most of the landslide points, but for areas with a low frequency of landslides, some models are insufficient. The CNN model demonstrates better results in the recognition of the landslides’ cluster region, this is also related to the convolution operation that takes the surrounding environment information into account. The higher accuracy and more concentrative possibility of CNN in LSA is of great significance for disaster prevention and mitigation, which can help the efficient use of human and material resources. Although CNN performs better than other methods, there are still some limitations, the identification of low-cluster landside areas can be enhanced by improving the CNN model.

Accurate modeling of the unresolved flame surface area is critical for the closure of reaction source terms in the flame surface density (FSD) method. Some algebraic models have been proposed for the unresolved flame surface area for premixed flames in the flamelet or thin reaction zones (TRZ) regimes where the Karlovitz number (Ka) is less than 100. However, in many lean combustion applications, Ka is large (Ka > 100) due to the strong interactions of small-scale turbulence and flames. In the present work, a direct numerical simulation (DNS) database was used to evaluate the performance of algebraic FSD models in high Ka premixed flames in the context of large eddy simulations. Three DNS cases, i.e., case L, case M and case H, were performed, where case L is located in the TRZ regime with Ka < 100 and case M and case H are located in the broken reaction zones regime with Ka > 100. A convolutional neural network (CNN) model was also developed to predict the generalized FSD, which was trained with samples of case H and a small filter size, and was tested in various cases with different Ka and filter sizes. It was found that the fraction of resolved FSD increases with increasing filtered progress variable c̃ and decreasing subgrid turbulent velocity fluctuation u′Δ. The performance of CNN and algebraic models was assessed using the DNS database. Overall, the results of algebraic models are promising in case L and case M for a small filter size; the CNN model performs generally better than the algebraic models in high Ka flames and the correlation coefficient between the modeled and actual generalized FSD is greater than 0.91 in all cases. The effects of c̃ and u′Δ on the performance of different models for various cases were explored. The algebraic models perform well with large values of c̃ and small values of u′Δ in high Ka cases, which indicates that they can be applied to high Ka flames in certain conditions. The performance of the CNN model is better than the algebraic models for a large filter size in high Ka cases.

To manage their disease, diabetic patients need to control the blood glucose level (BGL) by monitoring it and predicting its future values. This allows to avoid high or low BGL by taking recommended actions in advance. In this paper, we conduct a comparative study of two emerging deep learning techniques: Long-Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN) for one-step and multi-steps-ahead forecasting of the BGL based on Continuous Glucose Monitoring (CGM) data. The objectives are twofold: 1) Determining the best strategies of multi-steps-ahead forecasting (MSF) to fit the CNN and LSTM models respectively, and 2) Comparing the performances of the CNN and LSTM models for one-step and multi-steps prediction. Toward these objectives, we firstly conducted series of experiments of a CNN model through parameters selection to determine its best configuration. The LSTM model we used in the present study was developed and evaluated in an earlier work. Thereafter, five MSF strategies were developed and evaluated for the CNN and LSTM models using the Root-Mean-Square Error (RMSE) with an horizon of 30 min. To statistically assess the differences between the performances of CNN and LSTM models, we used the Wilcoxon statistical test. The results showed that: 1) no MSF strategy outperformed the others for both CNN and LSTM models, and 2) the proposed CNN model significantly outperformed the LSTM model for both one-step and multi-steps prediction.

Excessive withdrawal of groundwater for agricultural irrigation can cause seawater intrusion into coastal aquifers. Such a case will in turn results in deterioration of irrigation water quality. Determination of irrigation water quality with traditional methods is a time-consuming and costly process. However, machine learning algorithms can be useful tools for modeling and estimating groundwater quality used for irrigation water purposes. In this study, TDS, PS, SAR, and Cl parameters of groundwater were estimated with models based on EC and pH variables. For this purpose, prediction performances of two different deep learning methods (convolutional neural network (CNN) and deep neural network (DNN)) and two different classical machine learning (Random Forest (RF) and extreme gradient boosting (XGBoost)) methods were compared. In addition, predictive uncertainty of the models was determined by quantile regression (QR) analysis. Performance criteria and results of uncertainty analysis revealed that CNN (in testing phase, NSE = 0.95 for TDS, NSE = 0.96 for PS, NSE = 0.67 for SAR and NSE = 0.93 for CI) and DNN (in testing phase, NSE = 0.91 for TDS, NSE = 0.91 for PS, NSE = 0.57 for SAR and NSE = 0.94 for Cl) models had quite a close performance in estimation of TDS, PS, SAR, and Cl parameters and higher than the other two classical machine learning methods. As a result, the CNN model can be considered the best performing model in estimating all quality parameters due to the highest NSE and lowest RMSE values. In addition, the Taylor diagram showed that the values estimated using the CNN model had the highest correlation with the measured data. It was determined that the model with the lowest uncertainty based on the PICP statistics was DNN, followed by the CNN model. However, the CNN model has predicted outliers more accurately. Present findings proved that deep learning models could offer efficient tools for predicting irrigation water quality parameters.

Convolutional Neural Networks (CNNs) have demonstrated impressive performance in complex machine learning tasks such as classification and regression problems. A reliable neural network structure plays a decisive role in CNN studies. Through comparing and analyzing the structure of neural networks, a model structure for better visualization performance has been discovered, and such a method supports the development of deep learning research. These studies are of particular importance in end-to-end systems for autonomous driving to imitate human driving, where the interpretability of the system is limited. Because of the uncertainty of the ground truth, for the determination of human steering in an image, it is difficult to accurately compare the visualization performance of different CNN models or different visualization methods. For practical applications, however, an objective and quantitative measure for assessing visualization performance is necessary. Therefore, a method to evaluate the visualization performance of CNN models using a driver model instead of human drivers is proposed, to generate a data set which can be used to determine the decisional point (ground truth) in the input image. Then, an exclusive method is also put forth, to quantitatively calculate the relationship between the decisional point (ground truth) and the visualization results produced by CNN models. In this paper, five CNN models as an autonomous steering controller are designed based on PilotNet, and the visualization abilities of each CNN models is compared by three evaluation indicators. By comparing the visualization performance of five different CNN models, it is shown that the proposed method can successfully assess the visualization level of the CNN model.

BackgroundCalcareous nannofossils are minute microfossils widely present in marine strata. Their identification holds significant value in studies related to stratigraphic dating, paleo-environmental evolution, and paleoclimate reconstruction. However, the process of identifying these fossils is time consuming, and the discrepancies between the results obtained from different manual identification methods are substantial, hindering quantification efforts. Therefore, it is necessary to explore automated assisted identification of fossil species. This study mainly focused on 18 key fossil species from the Miocene era. Five convolutional neural network (CNN) models and 10 data augmentation techniques were compared. These models and techniques were employed to analyze and collectively train two- and three-dimensional fossil morphologies and structures obtained from three different fossils observed under single-polarized light microscopy, orthogonal polarized light microscopy, and scanning electron microscopy. Finally, the model performance was evaluated based on the predictive outcomes on the test set, using metrics such as confusion matrix and top-k accuracy. ResultThe results indicate that, for the calcareous nannofossil images, the most effective data augmentation approach is a combination of four methods: random rotation, random mirroring, random brightness, and gamma correction. Among the CNN models, DenseNet121 exhibits the optimal performance, achieving an identification accuracy of 94.56%. Moreover, this model can distinguish other fossils beyond the 18 key fossil species and non-fossil debris. Based on the confusion matrix, the evaluation results reveal that the model has strong generalization capability and outputs highly credible identification results.ConclusionDrawing on the identification results from CNN, this study asserts a robust correlation among extinction photographs, planar images, and stereoscopic morphological images of fossil species. Collective training facilitates the joint extraction and analysis of fossil features under different imaging methods. CNN demonstrates many advantages in the identification of calcareous nannofossils, offering convenience to researchers in various fields, such as stratigraphy, paleo-ecology, paleoclimate, and paleo-environments of ancient oceans. It has great potential for advancing the development of marine surveys and stratigraphic recognition processes in the future.

Development of Efficient CNN model for Tomato crop disease identification

This study proposes a method to extract the signature bands from the deep learning models of multispectral data converted from the hyperspectral data. The signature bands with two deep-learning models were further used to predict the sugar content of the Syzygium samarangense. Firstly, the hyperspectral data with the bandwidths lower than 2.5 nm were converted to the spectral data with multiple bandwidths higher than 2.5 nm to simulate the multispectral data. The convolution neural network (CNN) and the feedforward neural network (FNN) used these spectral data to predict the sugar content of the Syzygium samarangense and obtained the lowest mean absolute error (MAE) of 0.400° Brix and 0.408° Brix, respectively. Secondly, the absolute mean of the integrated gradient method was used to extract multiple signature bands from the CNN and FNN models for sugariness prediction. A total of thirty sets of six signature bands were selected from the CNN and FNN models, which were trained by using the spectral data with five bandwidths in the visible (VIS), visible to near-infrared (VISNIR), and visible to short-waved infrared (VISWIR) wavelengths ranging from 400 to 700 nm, 400 to 1000 nm, and 400 to 1700 nm. Lastly, these signature-band data were used to train the CNN and FNN models for sugar content prediction. The FNN model using VISWIR signature bands with a bandwidth of ± 12.5 nm had a minimum MAE of 0.390°Brix compared to the others. The CNN model using VISWIR signature bands with a bandwidth of ± 10 nm had the lowest MAE of 0.549° Brix compared to the other CNN models. The MAEs of the models with only six spectral bands were even better than those with tens or hundreds of spectral bands. These results reveal that six signature bands have the potential to be used in a small and compact multispectral device to predict the sugar content of the Syzygium samarangense.

Exploring biological motion perception in two-stream convolutional neural networks