Search

In the past years, several machine-learning-based techniques have arisen for providing effective crop protection. For instance, deep neural networks have been used to identify different types of weeds under different real-world conditions. However, these techniques usually require extensive involvement of experts working iteratively in the development of the most suitable machine learning system. To support this task and save resources, a new technique called Automated Machine Learning has started being studied. In this work, a complete open-source Automated Machine Learning system was evaluated with two different datasets, (i) The Early Crop Weeds dataset and (ii) the Plant Seedlings dataset, covering the weeds identification problem. Different configurations, such as the use of plant segmentation, the use of classifier ensembles instead of Softmax and training with noisy data, have been compared. The results showed promising performances of 93.8% and 90.74% F1 score depending on the dataset used. These performances were aligned with other related works in AutoML, but they are far from machine-learning-based systems manually fine-tuned by human experts. From these results, it can be concluded that finding a balance between manual expert work and Automated Machine Learning will be an interesting path to work in order to increase the efficiency in plant protection.

The effective detection of driver drowsiness is an important measure to prevent traffic accidents. Most existing drowsiness detection methods only use a single facial feature to identify fatigue status, ignoring the complex correlation between fatigue features and the time information of fatigue features, and this reduces the recognition accuracy. To solve these problems, we propose a driver sleepiness estimation model based on factorized bilinear feature fusion and a long- short-term recurrent convolutional network to detect driver drowsiness efficiently and accurately. The proposed framework includes three models: fatigue feature extraction, fatigue feature fusion, and driver drowsiness detection. First, we used a convolutional neural network (CNN) to effectively extract the deep representation of eye and mouth-related fatigue features from the face area detected in each video frame. Then, based on the factorized bilinear feature fusion model, we performed a nonlinear fusion of the deep feature representations of the eyes and mouth. Finally, we input a series of fused frame-level features into a long-short-term memory (LSTM) unit to obtain the time information of the features and used the softmax classifier to detect sleepiness. The proposed framework was evaluated with the National Tsing Hua University drowsy driver detection (NTHU-DDD) video dataset. The experimental results showed that this method had better stability and robustness compared with other methods.

Pedestrian gender classification is one of the key assignments of pedestrian study, and it finds practical applications in content-based image retrieval, population statistics, human–computer interaction, health care, multimedia retrieval systems, demographic collection, and visual surveillance. In this research work, gender classification was carried out using a deep learning approach. A new 64-layer architecture named 4-BSMAB derived from deep AlexNet is proposed. The proposed model was trained on CIFAR-100 dataset utilizing SoftMax classifier. Then, features were obtained from applied datasets with this pre-trained model. The obtained feature set was optimized with ant colony system (ACS) optimization technique. Various classifiers of SVM and KNN were used to perform gender classification utilizing the optimized feature set. Comprehensive experimentation was performed on gender classification datasets, and proposed model produced better results than the existing methods. The suggested model attained highest accuracy, i.e., 85.4%, and 92% AUC on MIT dataset, and best classification results, i.e., 93% accuracy and 96% AUC, on PKU-Reid dataset. The outcomes of extensive experiments carried out on existing standard pedestrian datasets demonstrate that the proposed framework outperformed existing pedestrian gender classification methods, and acceptable results prove the proposed model as a robust model.

Heart disease is the leading cause of death globally. The most common type of heart disease is coronary heart disease, which occurs when there is a build-up of plaque inside the arteries that supply blood to the heart, making blood circulation difficult. The prediction of heart disease is a challenge in clinical machine learning. Early detection of people at risk of the disease is vital in preventing its progression. This paper proposes a deep learning approach to achieve improved prediction of heart disease. An enhanced stacked sparse autoencoder network (SSAE) is developed to achieve efficient feature learning. The network consists of multiple sparse autoencoders and a softmax classifier. Additionally, in deep learning models, the algorithm’s parameters need to be optimized appropriately to obtain efficient performance. Hence, we propose a particle swarm optimization (PSO) based technique to tune the parameters of the stacked sparse autoencoder. The optimization by the PSO improves the feature learning and classification performance of the SSAE. Meanwhile, the multilayer architecture of autoencoders usually leads to internal covariate shift, a problem that affects the generalization ability of the network, hence, batch normalization is introduced to prevent this problem. The experimental results show that the proposed method effectively predicts heart disease by obtaining a classification accuracy of 0.973 and 0.961 on the Framingham and Cleveland heart disease datasets, respectively, thereby outperforming other machine learning methods and similar studies.

Extraction of coronary arteries in coronary computed tomography (CT) angiography is a prerequisite for the quantification of coronary lesions. In this study, we propose a tracking method combining a deep convolutional neural network (DNN) and particle filtering method to identify the trajectories from the coronary ostium to each distal end from 3D CT images. The particle filter, as a non-linear approximator, is an appropriate tracking framework for such thin and elongated structures, however, the robust `vesselness’ measurement is essential for extracting coronary centerlines. Importantly, we employed the DNN to robustly measure the vesselness using patch images, and we integrated softmax values to the likelihood function in our particle filtering framework. Tangent patches represent cross-sections of coronary arteries of circular shapes. Thus, 2D tangent patches are assumed to include enough features of coronary arteries, and the use of 2D patches significantly reduces computational complexity. Because coronary vasculature has multiple bifurcations, we also modeled a method to detect branching sites by clustering the particle locations. The proposed method is compared with three commercial workstations and two conventional methods from the academic literature.

How to deal with rare and unknown data in traffic classification has a decisive influence on classification performance. Rare data make it difficult to generate validation datasets to prevent overfitting, and unknown data interferes with learning and degrades the performance of the model. This paper presents a model generation method that accurately classifies rare data and new types of attacks, and does not result in overfitting. First, we use oversampling methods to solve the data imbalance caused by rare data. We separate the test dataset into a training dataset and a validation dataset. A model is created using separate training and validation datasets. Furthermore, the test dataset is used only for evaluating the performance capabilities of classification models, in order to make the test dataset independent of learning. We also use a softmax function that numerically indicates the probability that the model’s predictive results are accurate in detecting new, unknown attacks. Consequently, when applying the proposed method to the NSL_KDD dataset, the accuracy is 91.66%—an improvement of 6–16% compared to existing methods.

Fault diagnosis of mechanical equipment is mainly based on the contact measurement and analysis of vibration signals. In some special working conditions, the non-contact fault diagnosis method represented by the measurement of acoustic signals can make up for the lack of contact testing. However, its engineering application value is greatly restricted due to the low signal-to-noise ratio (SNR) of the acoustic signal. To solve this deficiency, a novel fault diagnosis method based on the generalized matrix norm sparse filtering (GMNSF) is proposed in this paper. Specially, the generalized matrix norm is introduced into the sparse filtering to seek the optimal sparse feature distribution to overcome the defect of low SNR of acoustic signals. Firstly, the collected acoustic signals are randomly overlapped to form the sample fragment data set. Then, three constraints are imposed on the multi-period data set by the GMNSF model to extract the sparse features in the sample. Finally, softmax is used to as a classifier to categorize different fault types. The diagnostic performance of the proposed method is verified by the bearing and planetary gear datasets. Results show that the GMNSF model has good feature extraction ability performance and anti-noise ability than other traditional methods.

In gastronomic tourism, food is viewed as the central tourist attraction. Specifically, indigenous food is known to represent the expression of local culture and identity. To promote gastronomic tourism, it is critical to have a model for the food business analytics system. This research undertakes an empirical evaluation of recent transfer learning models for deep learning feature extraction for a food recognition model. The VIREO-Food172 Dataset and a newly established Sabah Food Dataset are used to evaluate the food recognition model. Afterwards, the model is implemented into a web application system as an attempt to automate food recognition. In this model, a fully connected layer with 11 and 10 Softmax neurons is used as the classifier for food categories in both datasets. Six pre-trained Convolutional Neural Network (CNN) models are evaluated as the feature extractors to extract essential features from food images. From the evaluation, the research found that the EfficientNet feature extractor-based and CNN classifier achieved the highest classification accuracy of 94.01% on the Sabah Food Dataset and 86.57% on VIREO-Food172 Dataset. EFFNet as a feature representation outperformed Xception in terms of overall performance. However, Xception can be considered despite some accuracy performance drawback if computational speed and memory space usage are more important than performance.

Automatic modulation classification (AMC) is the premise for signal detection and demodulation applications, especially in non-cooperative communication scenarios. It has been a popular topic for decades and has gained significant progress with the development of deep learning methods. To further improve classification accuracy, a hierarchical multifeature fusion (HMF) based on a multidimensional convolutional neural network (CNN)-long short-term memory (LSTM) network is proposed in this paper. First, a multidimensional CNN module (MD-CNN) is proposed for feature compensation between interactive features extracted by two-dimensional convolutional filters and respective features extracted by one-dimensional filters. Second, learnt features of the MD-CNN module are fed into an LSTM layer for further exploitation of temporal features. Finally, classification results are obtained by the Softmax classifier. The effectiveness of the proposed method is verified by abundant experimental results on two public datasets, RadioML.2016.10a and RadioML.2016.10b. Satisfying results are obtained as compared with state-of-the-art methods.

Recently, digital pathology is an essential application for clinical practice and medical research. Due to the lack of large annotated datasets, the deep transfer learning technique is often used to classify histopathology images. A softmax classifier is often used to perform classification tasks. Besides, a Support Vector Machine (SVM) classifier is also popularly employed, especially for binary classification problems. Accurately determining the category of the histopathology images is vital for the diagnosis of diseases. In this paper, the conventional softmax classifier and the SVM classifier-based transfer learning approach are evaluated to classify histopathology cancer images in a binary breast cancer dataset and a multiclass lung and colon cancer dataset. In order to achieve better classification accuracy, a methodology that attaches SVM classifier to the fully-connected (FC) layer of the softmax-based transfer learning model is proposed. The proposed architecture involves a first step training the newly added FC layer on the target dataset using the softmax-based model and a second step training the SVM classifier with the newly trained FC layer. Cross-validation is used to ensure no bias for the evaluation of the performance of the models. Experimental results reveal that the conventional SVM classifier-based model is the least accurate on either binary or multiclass cancer datasets. The conventional softmax-based model shows moderate classification accuracy, while the proposed synthetic architecture achieves the best classification accuracy.

The convolutional neural network (CNN) has been proven to have better performance in hyperspectral image (HSI) classification than traditional methods. Traditional CNN on hyperspectral image classification is used to pay more attention to spectral features and ignore spatial information. In this paper, a new HSI model called local and hybrid dilated convolution fusion network (LDFN) was proposed, which fuses the local information of details and rich spatial features by expanding the perception field. The details of our local and hybrid dilated convolution fusion network methods are as follows. First, many operations are selected, such as standard convolution, average pooling, dropout and batch normalization. Then, fusion operations of local and hybrid dilated convolution are included to extract rich spatial-spectral information. Last, different convolution layers are gathered into residual fusion networks and finally input into the softmax layer to classify. Three widely hyperspectral datasets (i.e., Salinas, Pavia University and Indian Pines) have been used in the experiments, which show that LDFN outperforms state-of-art classifiers.

Unlike most of the existing neural network-based fall detection methods, which only detect fall at the time range, the algorithm proposed in this paper detect fall in both spatial and temporal dimension. A movement tube detection network integrating 3D CNN and object detection framework such as SSD is proposed to detect human fall with constrained movement tubes. The constrained movement tube, which encapsulates the person with a sequence of bounding boxes, has the merits of encapsulating the person closely and avoiding peripheral interference. A 3D convolutional neural network is used to encode the motion and appearance features of a video clip, which are fed into the tube anchors generation layer, softmax classification, and movement tube regression layer. The movement tube regression layer fine tunes the tube anchors to the constrained movement tubes. A large-scale spatio-temporal (LSST) fall dataset is constructed using self-collected data to evaluate the fall detection in both spatial and temporal dimensions. LSST has three characteristics of large scale, annotation, and posture and viewpoint diversities. Furthermore, the comparative experiments on a public dataset demonstrate that the proposed algorithm achieved sensitivity, specificity an accuracy of 100%, 97.04%, and 97.23%, respectively, outperforms the existing methods.

Credit scoring is a process of determining whether a borrower is successful or unsuccessful in repaying a loan using borrowers’ qualitative and quantitative characteristics. In recent years, machine learning algorithms have become widely studied in the development of credit scoring models. Although efficiently classifying good and bad borrowers is a core objective of the credit scoring model, there is still a need for the model that can explain the relationship between input and output. In this work, we propose a novel partially interpretable adaptive softmax (PIA-Soft) regression model to achieve both state-of-the-art predictive performance and marginally interpretation between input and output. We augment softmax regression by neural networks to make it adaptive for each borrower. Our PIA-Soft model consists of two main components: linear (softmax regression) and non-linear (neural network). The linear part explains the fundamental relationship between input and output variables. The non-linear part serves to improve the prediction performance by identifying the non-linear relationship between features for each borrower. The experimental result on public benchmark datasets shows that our proposed model not only outperformed the machine learning baselines but also showed the explanations that logically related to the real-world.

In 2013, China proposed the “Belt & Road Initiative” which aims to invest the “Belt & Road” countries so as to help them develop their infrastructure and economy. China consumes the largest part of fossil energy of the whole world, so it is China’s priority to consider its energy supplying security. Therefore, it becomes urgent for China to invest the “Belt & Road” countries’ energy facilities. There comes a question: how to evaluate the overseas energy investment risk? To answer this question, this paper proposes a deep learning method to assess such risk of the 50 “Belt & Road” countries. Specifically, this paper first proposes an indicator system in which 6 main factors are separated into 36 sub-factors. This paper makes use of hierarchical convolution neural networks (CNN) to encode the historical statistics. The hierarchical structure could help CNN handle the long historical statistics more effectively and efficiently. Afterward, this paper leverages the self-attention layer to calculate the weights of each sub-factor. It could be observed that the resource potential is the most important indicator, while “years of China’s diplomatic relations” is the most important sub-indicator. Finally, we use a conditional random field (CRF) layer and softmax layer to compute the assessment scores of each country. Based on the experimental results, this paper suggests Russia, United Arab Emirates (UAE), Malaysia, Saudi Arabia, Pakistan, Indonesia, and Kazakhstan to be China’s most reliable choices for energy investment.

To design an efficient deep learning model that can be used in the real-world, it is important to detect out-of-distribution (OOD) data well. Various studies have been conducted to solve the OOD problem. The current state-of-the-art approach uses a confidence score based on the Mahalanobis distance in a feature space. Although it outperformed the previous approaches, the results were sensitive to the quality of the trained model and the dataset complexity. Herein, we propose a novel OOD detection method that can train more efficient feature space for OOD detection. The proposed method uses an ensemble of the features trained using the softmax-based classifier and the network based on distance metric learning (DML). Through the complementary interaction of these two networks, the trained feature space has a more clumped distribution and can fit well on the Gaussian distribution by class. Therefore, OOD data can be efficiently detected by setting a threshold in the trained feature space. To evaluate the proposed method, we applied our method to various combinations of image datasets. The results show that the overall performance of the proposed approach is superior to those of other methods, including the state-of-the-art approach, on any combination of datasets.

Ship detection and tracking is an important task in video surveillance in inland waterways. However, ships in inland navigation are faced with accidents such as collisions. For collision avoidance, we should strengthen the monitoring of navigation and the robustness of the entire system. Hence, this paper presents ship detection and tracking of ships using the improved You Only Look Once version 3 (YOLOv3) detection algorithm and Deep Simple Online and Real-time Tracking (Deep SORT) tracking algorithm. Three improvements are made to the YOLOv3 target detection algorithm. Firstly, the Kmeans clustering algorithm is used to optimize the initial value of the anchor frame to make it more suitable for ship application scenarios. Secondly, the output classifier is modified to a single Softmax classifier to suit our ship dataset which has three ship categories and mutual exclusion. Finally, Soft Non-Maximum Suppression (Soft-NMS) is introduced to solve the deficiencies of the Non-Maximum Suppression (NMS) algorithm when screening candidate frames. Results showed the mean Average Precision (mAP) and Frame Per Second (FPS) of the improved algorithm are increased by about 5% and 2, respectively, compared with the existing YOLOv3 detecting Algorithm. Then the improved YOLOv3 is applied in Deep Sort and the performance result of Deep Sort showed that, it has greater performance in complex scenes, and is robust to interference such as occlusion and camera movement, compared to state of art algorithms such as KCF, MIL, MOSSE, TLD, and Median Flow. With this improvement, it will help in the safety of inland navigation and protection from collisions and accidents.

The automatic modulation classification (AMC) of a detected signal has gained considerable prominence in recent years owing to its numerous facilities. Numerous studies have focused on feature-based AMC. However, improving accuracy under low signal-to-noise ratio (SNR) rates is a serious issue in AMC. Moreover, research on the enhancement of AMC performance under low and high SNR rates is limited. Motivated by these issues, this study proposes AMC using a feature clustering-based two-lane capsule network (AMC2N). In the AMC2N, accuracy of the MC process is improved by designing a new two-layer capsule network (TL-CapsNet), and classification time is reduced by introducing a new feature clustering approach in the TL-CapsNet. Firstly, the AMC2N executes blind equalization, sampling, and quantization in trilevel preprocessing. Blind equalization is executed using a binary constant modulus algorithm to avoid intersymbol interference. To extract features from the preprocessed signal and classify signals accurately, the AMC2N employs the TL-CapsNet, in which individual lanes are incorporated to process the real and imaginary parts of the signal. In addition, it is robust to SNR variations, that is, low and high SNR rates. The TL-CapsNet extracts features from the real and imaginary parts of the given signal, which are then clustered based on feature similarity. For feature extraction and clustering, the dynamic routing procedure of the TL-CapsNet is adopted. Finally, classification is performed in the SoftMax layer of the TL-CapsNet. This study proves that the AMC2N outperforms existing methods, particularly, convolutional neural network(CNN), Robust-CNN (R-CNN), curriculum learning(CL), and Local Binary Pattern (LBP), in terms of accuracy, precision, recall, F-score, and computation time. All metrics are validated in two scenarios, and the proposed method shows promising results in both.

The emergence of a large number of new malicious code poses a serious threat to network security, and most of them are derivative versions of existing malicious code. The classification of malicious code is helpful to analyze the evolutionary trend of malicious code families and trace the source of cybercrime. The existing methods of malware classification emphasize the depth of the neural network, which has the problems of a long training time and large computational cost. In this work, we propose the shallow neural network-based malware classifier (SNNMAC), a malware classification model based on shallow neural networks and static analysis. Our approach bridges the gap between precise but slow methods and fast but less precise methods in existing works. For each sample, we first generate n-grams from their opcode sequences of the binary file with a decompiler. An improved n-gram algorithm based on control transfer instructions is designed to reduce the n-gram dataset. Then, the SNNMAC exploits a shallow neural network, replacing the full connection layer and softmax with the average pooling layer and hierarchical softmax, to learn from the dataset and perform classification. We perform experiments on the Microsoft malware dataset. The evaluation result shows that the SNNMAC outperforms most of the related works with 99.21% classification precision and reduces the training time by more than half when compared with the methods using DNN (Deep Neural Networks).

The physical features of fruit and vegetables make the task of vision-based classification of fruit and vegetables challenging. The classification of fruit and vegetables at a supermarket self-checkout poses even more challenges due to variable lighting conditions and human factors arising from customer interactions with the system along with the challenges associated with the colour, texture, shape, and size of a fruit or vegetable. Considering this complex application, we have proposed a progressive coarse to fine classification technique to classify fruit and vegetables at supermarket checkouts. The image and weight of fruit and vegetables have been obtained using a prototype designed to simulate the supermarket environment, including the lighting conditions. The weight information is used to change the coarse classification of 15 classes down to three, which are further used in AdaBoost-based Convolutional Neural Network (CNN) optimisation for fine classification. The training samples for each coarse class are weighted based on AdaBoost optimisation, which are updated on each iteration of a training phase. Multi-class likelihood distribution obtained by the fine classification stage is used to estimate a final classification with a softmax classifier. GoogleNet, MobileNet, and a custom CNN have been used for AdaBoost optimisation, with promising classification results.

Visible image quality is very susceptible to changes in illumination, and there are limitations in ship classification using images acquired by a single sensor. This study proposes a ship classification method based on an attention mechanism and multi-scale convolutional neural network (MSCNN) for visible and infrared images. First, the features of visible and infrared images are extracted by a two-stream symmetric multi-scale convolutional neural network module, and then concatenated to make full use of the complementary features present in multi-modal images. After that, the attention mechanism is applied to the concatenated fusion features to emphasize local details areas in the feature map, aiming to further improve feature representation capability of the model. Lastly, attention weights and the original concatenated fusion features are added element by element and fed into fully connected layers and Softmax output layer for final classification output. Effectiveness of the proposed method is verified on a visible and infrared spectra (VAIS) dataset, which shows 93.81% accuracy in classification results. Compared with other state-of-the-art methods, the proposed method could extract features more effectively and has better overall classification performance.

The performance of a long short-term memory (LSTM) recurrent neural network (RNN)-based language model has been improved on language model benchmarks. Although a recurrent layer has been widely used, previous studies showed that an LSTM RNN-based language model (LM) cannot overcome the limitation of the context length. To train LMs on longer sequences, attention mechanism-based models have recently been used. In this paper, we propose a LM using a neural Turing machine (NTM) architecture based on localized content-based addressing (LCA). The NTM architecture is one of the attention-based model. However, the NTM encounters a problem with content-based addressing because all memory addresses need to be accessed for calculating cosine similarities. To address this problem, we propose an LCA method. The LCA method searches for the maximum of all cosine similarities generated from all memory addresses. Next, a specific memory area including the selected memory address is normalized with the softmax function. The LCA method is applied to pre-trained NTM-based LM during the test stage. The proposed architecture is evaluated on Penn Treebank and enwik8 LM tasks. The experimental results indicate that the proposed approach outperforms the previous NTM architecture.

Internet of Things (IoT) devices bring us rich sensor data, such as images capturing the environment. One prominent approach to understanding and utilizing such data is image classification which can be effectively solved by deep learning (DL). Combined with cross-entropy loss, softmax has been widely used for classification problems, despite its limitations. Many efforts have been made to enhance the performance of softmax decision-making models. However, they require complex computations and/or re-training the model, which is computationally prohibited on low-power IoT devices. In this paper, we propose a light-weight framework to enhance the performance of softmax decision-making models for DL. The proposed framework operates with a pre-trained DL model using softmax, without requiring any modification to the model. First, it computes the level of uncertainty as to the model’s prediction, with which misclassified samples are detected. Then, it makes a probabilistic control decision to enhance the decision performance of the given model. We validated the proposed framework by conducting an experiment for IoT car control. The proposed model successfully reduced the control decision errors by up to 96.77% compared to the given DL model, and that suggests the feasibility of building DL-based IoT applications with high accuracy and low complexity.

Convolutional neural networks (CNNs) have made great achievements on computer vision tasks, especially the image classification. With the improvement of network structure and loss functions, the performance of image classification is getting higher and higher. The classic Softmax + cross-entropy loss has been the norm for training neural networks for years, which is calculated from the output probability of the ground-truth class. Then the network&rsquo, s weight is updated by gradient calculation of the loss. However, after several epochs of training, the back-propagation errors usually become almost negligible. For the above considerations, we proposed that batch normalization with adjustable scale could be added after network output to alleviate the problem of vanishing gradient problem in deep learning. The experimental results show that our method can significantly improve the final classification accuracy on different network structures, and is also better than many other improved classification Loss.

Intelligent methods have long been researched in fault diagnosis. Traditionally, feature extraction and fault classification are separated, and this process is not completely intelligent. In addition, most traditional intelligent methods use an individual model, which cannot extract the discriminate features when the machines work in a complex condition. To overcome the shortcomings of traditional intelligent fault diagnosis methods, in this paper, an intelligent bearing fault diagnosis method based on ensemble sparse auto-encoders was proposed. Three different sparse auto-encoders were used as the main architecture. To improve the robustness and stability, a novel weight strategy based on distance metric and standard deviation metric was employed to assign the weights of three sparse auto-encodes. Softmax classifier is used to classify the fault types of integrated features. The effectiveness of the proposed method is validated with extensive experiments, and comparisons with the related methods and researches on the widely-used motor bearing dataset verify the superiority of the proposed method. The results show that the testing accuracy and the standard deviation are 99.71% and 0.05%.

Person re-identification (Re-ID) is a non-overlapping multi-camera retrieval task to match different images of the same person, and it has become a hot research topic in many fields, such as surveillance security, criminal investigation, and video analysis. As one kind of important architecture for person re-identification, Siamese networks usually adopt standard softmax loss function, and they can only obtain the global features of person images, ignoring the local features and the large margin for classification. In this paper, we design a novel symmetric Siamese network model named Siamese Multiple Granularity Network (SMGN), which can jointly learn the large margin multiple granularity features and similarity metrics for person re-identification. Firstly, two branches for global and local feature extraction are designed in the backbone of the proposed SMGN model, and the extracted features are concatenated together as multiple granularity features of person images. Then, to enhance their discriminating ability, the multiple channel weighted fusion (MCWF) loss function is constructed for the SMGN model, which includes the verification loss and identification loss of the training image pair. Extensive comparative experiments on four benchmark datasets (CUHK01, CUHK03, Market-1501 and DukeMTMC-reID) show the effectiveness of our proposed method and its performance outperforms many state-of-the-art methods.

A rich line of works focus on designing elegant loss functions under the deep metric learning (DML) paradigm to learn a discriminative embedding space for remote sensing image retrieval (RSIR). Essentially, such embedding space could efficiently distinguish deep feature descriptors. So far, most existing losses used in RSIR are based on triplets, which have disadvantages of local optimization, slow convergence and insufficient use of similarity structure in a mini-batch. In this paper, we present a novel DML method named as global optimal structured loss to deal with the limitation of triplet loss. To be specific, we use a softmax function rather than a hinge function in our novel loss to realize global optimization. In addition, we present a novel optimal structured loss, which globally learn an efficient deep embedding space with mined informative sample pairs to force the positive pairs within a limitation and push the negative ones far away from a given boundary. We have conducted extensive experiments on four public remote sensing datasets and the results show that the proposed global optimal structured loss with pairs mining scheme achieves the state-of-the-art performance compared with the baselines.

Speech is the most significant mode of communication among human beings and a potential method for human-computer interaction (HCI) by using a microphone sensor. Quantifiable emotion recognition using these sensors from speech signals is an emerging area of research in HCI, which applies to multiple applications such as human-reboot interaction, virtual reality, behavior assessment, healthcare, and emergency call centers to determine the speaker’s emotional state from an individual’s speech. In this paper, we present major contributions for, (i) increasing the accuracy of speech emotion recognition (SER) compared to state of the art and (ii) reducing the computational complexity of the presented SER model. We propose an artificial intelligence-assisted deep stride convolutional neural network (DSCNN) architecture using the plain nets strategy to learn salient and discriminative features from spectrogram of speech signals that are enhanced in prior steps to perform better. Local hidden patterns are learned in convolutional layers with special strides to down-sample the feature maps rather than pooling layer and global discriminative features are learned in fully connected layers. A SoftMax classifier is used for the classification of emotions in speech. The proposed technique is evaluated on Interactive Emotional Dyadic Motion Capture (IEMOCAP) and Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) datasets to improve accuracy by 7.85% and 4.5%, respectively, with the model size reduced by 34.5 MB. It proves the effectiveness and significance of the proposed SER technique and reveals its applicability in real-world applications.

Advancements in artificial intelligence (AI) gives a great opportunity to develop an autonomous devices. The contribution of this work is an improved convolutional neural network (CNN) model and its implementation for the detection of road cracks, potholes, and yellow lane in the road. The purpose of yellow lane detection and tracking is to realize autonomous navigation of unmanned aerial vehicle (UAV) by following yellow lane while detecting and reporting the road cracks and potholes to the server through WIFI or 5G medium. The fabrication of own data set is a hectic and time-consuming task. The data set is created, labeled and trained using default and an improved model. The performance of both these models is benchmarked with respect to accuracy, mean average precision (mAP) and detection time. In the testing phase, it was observed that the performance of the improved model is better in respect of accuracy and mAP. The improved model is implemented in UAV using the robot operating system for the autonomous detection of potholes and cracks in roads via UAV front camera vision in real-time., Comment: 10 pages, 11 figures, journal

Wheat is a very important food crop for mankind. Many new varieties are bred every year. The accurate judgment of wheat varieties can promote the development of the wheat industry and the protection of breeding property rights. Although gene analysis technology can be used to accurately determine wheat varieties, it is costly, time-consuming, and inconvenient. Traditional machine learning methods can significantly reduce the cost and time of wheat cultivars identification, but the accuracy is not high. In recent years, the relatively popular deep learning methods have further improved the accuracy on the basis of traditional machine learning, whereas it is quite difficult to continue to improve the identification accuracy after the convergence of the deep learning model. Based on the ResNet and SENet models, this paper draws on the idea of the bagging-based ensemble estimator algorithm, and proposes a deep learning model for wheat classification, CMPNet, which is coupled with the tillering period, flowering period, and seed image. This convolutional neural network (CNN) model has a symmetrical structure along the direction of the tensor flow. The model uses collected images of different types of wheat in multiple growth periods. First, it uses the transfer learning method of the ResNet-50, SE-ResNet, and SE-ResNeXt models, and then trains the collected images of 30 kinds of wheat in different growth periods. It then uses the concat layer to connect the output layers of the three models, and finally obtains the wheat classification results through the softmax function. The accuracy of wheat variety identification increased from 92.07% at the seed stage, 95.16% at the tillering stage, and 97.38% at the flowering stage to 99.51%. The model’s single inference time was only 0.0212 s. The model not only significantly improves the classification accuracy of wheat varieties, but also achieves low cost and high efficiency, which makes it a novel and important technology reference for wheat producers, managers, and law enforcement supervisors in the practice of wheat production.

This paper develops a novel hybrid feature learner and classifier for vibration-based fault detection and isolation (FDI) of industrial apartments. The trained model extracts high-level discriminative features from vibration signals and predicts equipment state. Against the limitations of traditional machine learning (ML)-based classifiers, the convolutional neural network (CNN) and deep neural network (DNN) are not only superior for real-time applications, but they also come with other benefits including ease-of-use, automated feature learning, and higher predictive accuracies. This study proposes a hybrid DNN and one-dimensional CNN diagnostics model (D-dCNN) which automatically extracts high-level discriminative features from vibration signals for FDI. Via Softmax averaging at the output layer, the model mitigates the limitations of the standalone classifiers. A diagnostic case study demonstrates the efficiency of the model with a significant accuracy of 92% (F1 score) and extensive comparative empirical validations.

In this research, we applied deep learning to rank the effectiveness of candidate drug compounds in combating viral cells, in particular, SARS-Cov-2 viral cells. For this purpose, two different datasets from Recursion Pharmaceuticals, a siRNA image dataset (RxRx1), which were used to build and calibrate our model for feature extraction, and a SARS-CoV-2 dataset (RxRx19a) was used to train our model for ranking efficacy of candidate drug compounds. The SARS-CoV-2 dataset contained healthy, uninfected control or “mock” cells, as well as “active viral” cells (cells infected with COVID-19), which were the two cell types used to train our deep learning model. In addition, it contains viral cells treated with different drug compounds, which were the cells not used to train but test our model. We devised a new cascade transfer learning strategy to construct our model. We first trained a deep learning model, the DenseNet, with the siRNA set, a dataset with characteristics similar to the SARS-CoV-2 dataset, for feature extraction. We then added additional layers, including a SoftMax layer as an output layer, and retrained the model with active viral cells and mock cells from the SARS-CoV-2 dataset. In the test phase, the SoftMax layer outputs probability (equivalently, efficacy) scores which allows us to rank candidate compounds, and to study the performance of each candidate compound statistically. With this approach, we identified several compounds with high efficacy scores which are promising for the therapeutic treatment of COVID-19. The compounds showing the most promise were GS-441524 and then Remdesivir, which overlapped with these reported in the literature and with these drugs that are approved by FDA, or going through clinical trials and preclinical trials. This study shows the potential of deep learning in its ability to identify promising compounds to aid rapid responses to future pandemic outbreaks.

Deepfake is one of the applications that is deemed harmful. Deepfakes are a sort of image or video manipulation in which a person’s image is changed or swapped with that of another person’s face using artificial neural networks. Deepfake manipulations may be done with a variety of techniques and applications. A quintessential countermeasure against deepfake or face manipulation is deepfake detection method. Most of the existing detection methods perform well under symmetric data distributions, but are still not robust to asymmetric datasets variations and novel deepfake/manipulation types. In this paper, for the identification of fake faces in videos, a new multistream deep learning algorithm is developed, where three streams are merged at the feature level using the fusion layer. After the fusion layer, the fully connected, Softmax, and classification layers are used to classify the data. The pre-trained VGG16 model is adopted for transferred CNN1stream. In transfer learning, the weights of the pre-trained CNN model are further used for training the new classification problem. In the second stream (transferred CNN2), the pre-trained VGG19 model is used. Whereas, in the third stream, the pre-trained ResNet18 model is considered. In this paper, a new large-scale dataset (i.e., World Politicians Deepfake Dataset (WPDD)) is introduced to improve deepfake detection systems. The dataset was created by downloading videos of 20 different politicians from YouTube. Over 320,000 frames were retrieved after dividing the downloaded movie into little sections and extracting the frames. Finally, various manipulations were performed to these frames, resulting in seven separate manipulation classes for men and women. In the experiments, three fake face detection scenarios are investigated. First, fake and real face discrimination is studied. Second, seven face manipulations are performed, including age, beard, face swap, glasses, hair color, hairstyle, smiling, and genuine face discrimination. Third, performance of deepfake detection system under novel type of face manipulation is analyzed. The proposed strategy outperforms the prior existing methods. The calculated performance metrics are over 99%.

In order to adequately characterize the visual characteristics of atmospheric visibility and overcome the disadvantages of the traditional atmospheric visibility measurement method with significant dependence on preset reference objects, high cost, and complicated steps, this paper proposed an ensemble learning method for atmospheric visibility grading based on deep neural network and stochastic weight averaging. An experiment was conducted using the scene of an expressway, and three visibility levels were set, i.e., Level 1, Level 2, and Level 3. Firstly, the EfficientNet was transferred to extract the abstract features of the images. Then, training and grading were performed on the feature sets through the SoftMax regression model. Subsequently, the feature sets were ensembled using the method of stochastic weight averaging to obtain the atmospheric visibility grading model. The obtained datasets were input into the grading model and tested. The grading model classified the results into three categories, with the grading accuracy being 95.00%, 89.45%, and 90.91%, respectively, and the average accuracy of 91.79%. The results obtained by the proposed method were compared with those obtained by the existing methods, and the proposed method showed better performance than those of other methods. This method can be used to classify the atmospheric visibility of traffic and reduce the incidence of traffic accidents caused by atmospheric visibility.

Design: At the heart of time series forecasting, if nonlinear and nonstationary data are analyzed using traditional time series, the results will be biased. At the same time, if just using machine learning without any consideration given to input from traditional time series, not much information can be obtained from the results because the machine learning model is a black box. Purpose: In order to better study time series forecasting, we extend the combination of traditional time series and machine learning and propose a hybrid cascade neural network considering a metaheuristic optimization genetic algorithm in space–time forecasting. Finding: To further show the utility of the cascade neural network genetic algorithm, we use various scenarios for training and testing while also extending simulations by considering the activation functions SoftMax, radbas, logsig, and tribas on space–time forecasting of pollution data. During the simulation, we perform numerical metric evaluations using the root-mean-square error (RMSE), mean absolute error (MAE), and symmetric mean absolute percentage error (sMAPE) to demonstrate that our models provide high accuracy and speed up time-lapse computing.

Circular RNAs (circRNAs) are extensively expressed in cells and tissues, and play crucial roles in human diseases and biological processes. Recent studies have reported that circRNAs could function as RNA binding protein (RBP) sponges, meanwhile RBPs can also be involved in back-splicing. The interaction with RBPs is also considered an important factor for investigating the function of circRNAs. Hence, it is necessary to understand the interaction mechanisms of circRNAs and RBPs, especially in human cancers. Here, we present a novel method based on deep learning to identify cancer-specific circRNA&ndash, RBP binding sites (CSCRSites), only using the nucleotide sequences as the input. In CSCRSites, an architecture with multiple convolution layers is utilized to detect the features of the raw circRNA sequence fragments, and further identify the binding sites through a fully connected layer with the softmax output. The experimental results show that CSCRSites outperform the conventional machine learning classifiers and some representative deep learning methods on the benchmark data. In addition, the features learnt by CSCRSites are converted to sequence motifs, some of which can match to human known RNA motifs involved in human diseases, especially cancer. Therefore, as a deep learning-based tool, CSCRSites could significantly contribute to the function analysis of cancer-associated circRNAs.

To monitor the tool wear state of computerized numerical control (CNC) machining equipment in real time in a manufacturing workshop, this paper proposes a real-time monitoring method based on a fusion of a convolutional neural network (CNN) and a bidirectional long short-term memory (BiLSTM) network with an attention mechanism (CABLSTM). In this method, the CNN is used to extract deep features from the time-series signal as an input, and then the BiLSTM network with a symmetric structure is constructed to learn the time-series information between the feature vectors. The attention mechanism is introduced to self-adaptively perceive the network weights associated with the classification results of the wear state and distribute the weights reasonably. Finally, the signal features of different weights are sent to a Softmax classifier to classify the tool wear state. In addition, a data acquisition experiment platform is developed with a high-precision CNC milling machine and an acceleration sensor to collect the vibration signals generated during tool processing in real time. The original data are directly fed into the depth neural network of the model for analysis, which avoids the complexity and limitations caused by a manual feature extraction. The experimental results show that, compared with other deep learning neural networks and traditional machine learning network models, the model can predict the tool wear state accurately in real time from original data collected by sensors, and the recognition accuracy and generalization have been improved to a certain extent.

This study improved on previous efforts to map longleaf pine (Pinus palustris) over large areas in the southeastern United States of America by developing new methods that integrate forest inventory data, aerial photography and Landsat 8 imagery to model forest characteristics. Spatial, statistical and machine learning algorithms were used to relate United States Forest Service Forest Inventory and Analysis (FIA) field plot data to relatively normalized Landsat 8 imagery based texture. Modeling algorithms employed include softmax neural networks and multiple hurdle models that combine softmax neural network predictions with linear regression models to estimate key forest characteristics across 2.3 million ha in Georgia, USA. Forest metrics include forest type, basal area and stand density. Results show strong relationships between Landsat 8 imagery based texture and field data (map accuracy > 0.80; square root basal area per ha residual standard errors < 1; natural log transformed trees per ha < 1.081). Model estimates depicting spatially explicit, fine resolution raster surfaces of forest characteristics for multiple coniferous and deciduous species across the study area were created and made available to the public in an online raster database. These products can be integrated with existing tabular, vector and raster databases already being used to guide longleaf pine conservation and restoration in the region.

In civil engineering, the stability of concrete is of great significance to safety of people&rsquo, s life and property, so it is necessary to detect concrete damage effectively. In this paper, we treat crack detection on concrete surface as a semantic segmentation task that distinguishes background from crack at the pixel level. Inspired by Fully Convolutional Networks (FCN), we propose a full convolution network based on dilated convolution for concrete crack detection, which consists of an encoder and a decoder. Specifically, we first used the residual network to extract the feature maps of the input image, designed the dilated convolutions with different dilation rates to extract the feature maps of different receptive fields, and fused the extracted features from multiple branches. Then, we exploited the stacked deconvolution to do up-sampling operator in the fused feature maps. Finally, we used the SoftMax function to classify the feature maps at the pixel level. In order to verify the validity of the model, we introduced the commonly used evaluation indicators of semantic segmentation: Pixel Accuracy (PA), Mean Pixel Accuracy (MPA), Mean Intersection over Union (MIoU), and Frequency Weighted Intersection over Union (FWIoU). The experimental results show that the proposed model converges faster and has better generalization performance on the test set by introducing dilated convolutions with different dilation rates and a multi-branch fusion strategy. Our model has a PA of 96.84%, MPA of 92.55%, MIoU of 86.05% and FWIoU of 94.22% on the test set, which is superior to other models.

For good performance, most existing electrocardiogram (ECG) identification methods still need to adopt a denoising process to remove noise interference beforehand. This specific signal preprocessing technique requires great efforts for algorithm engineering and is usually complicated and time-consuming. To more conveniently remove the influence of noise interference and realize accurate identification, a novel temporal-frequency autoencoding based method is proposed. In particular, the raw data is firstly transformed into the wavelet domain, where multi-level time-frequency representation is achieved. Then, a prior knowledge-based feature selection is proposed and applied to the transformed data to discard noise components and retain identity-related information simultaneously. Afterward, the stacked sparse autoencoder is introduced to learn intrinsic discriminative features from the selected data, and Softmax classifier is used to perform the identification task. The effectiveness of the proposed method is evaluated on two public databases, namely, ECG-ID and Massachusetts Institute of Technology-Biotechnology arrhythmia (MIT-BIH-AHA) databases. Experimental results show that our method can achieve high multiple-heartbeat identification accuracies of 98.87%, 92.3%, and 96.82% on raw ECG signals which are from the ECG-ID (Two-recording), ECG-ID (All-recording), and MIT-BIH-AHA database, respectively, indicating that our method can provide an efficient way for ECG biometric identification.

As the demand for over-the-top and online streaming services exponentially increases, many techniques for Quality of Experience (QoE) provisioning have been studied. Users can take actions (e.g., skipping) while streaming a video. Therefore, we should consider the viewing pattern of users rather than the network condition or video quality. In this context, we propose a proactive content-loading algorithm for improving per-user personalized preferences using multinomial softmax classification. Based on experimental results, the proposed algorithm has a personalized per-user content waiting time that is significantly lower than that of competing algorithms.

Accurate and timely bearing fault diagnosis is crucial to decrease the probability of unexpected failures of rotating machinery and improve the efficiency of its scheduled maintenance. Since convolutional neural networks (CNN) have poor feature extraction capability for sensor data with 1D format, CNN combined with signal processing algorithm is often adopted for fault diagnosis. This increases manual conversion work and expertise dependence while reducing the feasibility and robustness of the corresponding fault diagnosis method. In this paper, a novel sensor data-driven fault diagnosis method is proposed by fusing S-transform (ST) algorithm and CNN, namely ST-CNN. First of all, a ST layer is designed based on S-transform algorithm. In the ST layer, sensor data is automatically converted into 2D time-frequency matrix without manual conversion work. Then, a new ST-CNN model is constructed, and the time-frequency coefficient matrixes are inputted into the constructed ST-CNN model. After the training process of the ST-CNN model is completed, the classification layer such as softmax performs the fault diagnosis. Finally, the diagnosis performance of the proposed method is evaluated by using two public available datasets of bearings. The experimental results show that the proposed method performs the higher and more robust diagnosis performance than other existing methods.

Polarimetric synthetic aperture radar (PolSAR) has become increasingly popular in the past two decades, for it can derive multichannel features of ground objects, which contains more discriminative information compared with traditional SAR. In this paper, a neural nonlocal stacked sparse autoencoders with virtual adversarial regularization (NNSSAE-VAT) is proposed for PolSAR image classification. The NNSSAE first extracts the nonlocal features by calculating pairwise similarity of each pixel and its surrounding pixels using a neural network, which contains a multiscale feature extractor and a linear embedding layer. The feature extraction process can relieve the negative influence of speckle noise and extract discriminative nonlocal spatial information without carefully designed parameters. Then, the SSAE maps the center pixel and the extracted nonlocal features into deep latent space in which a Softmax classifier is utilized to conduct classification. The virtual adversarial training is introduced to regularize the network, which tries to keep the network from being overfitting. The experimental results from three real PolSAR image show that the proposed NNSSAE-VAT method has proved its robustness and effectiveness and it can achieve competitive performance compared with related methods.

The worldwide utilization of surveillance cameras in smart cities has enabled researchers to analyze a gigantic volume of data to ensure automatic monitoring. An enhanced security system in smart cities, schools, hospitals, and other surveillance domains is mandatory for the detection of violent or abnormal activities to avoid any casualties which could cause social, economic, and ecological damages. Automatic detection of violence for quick actions is very significant and can efficiently assist the concerned departments. In this paper, we propose a triple-staged end-to-end deep learning violence detection framework. First, persons are detected in the surveillance video stream using a light-weight convolutional neural network (CNN) model to reduce and overcome the voluminous processing of useless frames. Second, a sequence of 16 frames with detected persons is passed to 3D CNN, where the spatiotemporal features of these sequences are extracted and fed to the Softmax classifier. Furthermore, we optimized the 3D CNN model using an open visual inference and neural networks optimization toolkit developed by Intel, which converts the trained model into intermediate representation and adjusts it for optimal execution at the end platform for the final prediction of violent activity. After detection of a violent activity, an alert is transmitted to the nearest police station or security department to take prompt preventive actions. We found that our proposed method outperforms the existing state-of-the-art methods for different benchmark datasets.

Lung cancer is the most common cause of cancer-related deaths worldwide. Hence, the survival rate of patients can be increased by early diagnosis. Recently, machine learning methods on Computed Tomography (CT) images have been used in the diagnosis of lung cancer to accelerate the diagnosis process and assist physicians. However, in conventional machine learning techniques, using handcrafted feature extraction methods on CT images are complicated processes. Hence, deep learning as an effective area of machine learning methods by using automatic feature extraction methods could minimize the process of feature extraction. In this study, two Convolutional Neural Network (CNN)-based models were proposed as deep learning methods to diagnose lung cancer on lung CT images. To investigate the performance of the two proposed models (Straight 3D-CNN with conventional softmax and hybrid 3D-CNN with Radial Basis Function (RBF)-based SVM), the altered models of two-well known CNN architectures (3D-AlexNet and 3D-GoogleNet) were considered. Experimental results showed that the performance of the two proposed models surpassed 3D-AlexNet and 3D-GoogleNet. Furthermore, the proposed hybrid 3D-CNN with SVM achieved more satisfying results (91.81%, 88.53% and 91.91% for accuracy rate, sensitivity and precision respectively) compared to straight 3D-CNN with softmax in the diagnosis of lung cancer.

The development and application of marine current energy are attracting more and more attention around the world. Due to the hardness of its working environment, it is important and difficult to study the fault diagnosis of a marine current generation system. In this paper, an underwater image is chosen as the fault-diagnosing signal, after different sensors are compared. This paper proposes a diagnosis method based on the sparse autoencoder (SA) and softmax regression (SR). The SA is used to extract the features and SR is used to classify them. Images are used to monitor whether the blade is attached by benthos and to determine its corresponding degree of attachment. Compared with other methods, the experiment results show that the proposed method can diagnose the blade attachment with higher accuracy.

One of the challenges in fine-grained classification is that subcategories with significant similarity are hard to be distinguished due to the equal treatment of all subcategories in existing algorithms. In order to solve this problem, a fine-grained image classification method by combining a bilinear convolutional neural network (B-CNN) and the measurement of subcategory similarities is proposed. Firstly, an improved weakly supervised localization method is designed to obtain the bounding box of the main object, which allows the model to eliminate the influence of background noise and obtain more accurate features. Then, sample features in the training set are computed by B-CNN so that the fuzzing similarity matrix for measuring interclass similarities can be obtained. To further improve classification accuracy, the loss function is designed by weighting triplet loss and softmax loss. Extensive experiments implemented on two benchmarks datasets, Stanford Cars-196 and Caltech-UCSD Birds-200-2011 (CUB-200-2011), show that the newly proposed method outperforms in accuracy several state-of-the-art weakly supervised classification models.

One-hot encoding is the prevalent method used in neural networks to represent multi-class categorical data. Its success stems from its ease of use and interpretability as a probability distribution when accompanied by a softmax activation function. However, one-hot encoding leads to very high dimensional vector representations when the categorical data’s cardinality is high. The Hamming distance in one-hot encoding is equal to two from the coding theory perspective, which does not allow detection or error-correcting capabilities. Binary coding provides more possibilities for encoding categorical data into the output codes, which mitigates the limitations of the one-hot encoding mentioned above. We propose a novel method based on Zadeh fuzzy logic to train binary output codes holistically. We study linear block codes for their possibility of separating class information from the checksum part of the codeword, showing their ability not only to detect recognition errors by calculating non-zero syndrome, but also to evaluate the truth-value of the decision. Experimental results show that the proposed approach achieves similar results as one-hot encoding with a softmax function in terms of accuracy, reliability, and out-of-distribution performance. It suggests a good foundation for future applications, mainly classification tasks with a high number of classes.

Mission-critical applications that rely on deep learning (DL) for automation suffer because DL models struggle to provide reliable indicators of failure. Reliable failure prediction can greatly improve the efficiency of a system, because it becomes easier to predict when human intervention is required. DL-based systems thus stand to benefit greatly from robust measures of uncertainty over model predictions. Monte Carlo dropout (MCD), a Bayesian method, and deep ensembles (DE) have emerged as two of the most popular and competitive ways to perform uncertainty estimation. Although literature exploring the usefulness of these approaches exists in medical imaging, robotics and autonomous driving domains, it is scarce to non-existent for remote sensing, and in particular, synthetic aperture radar (SAR) applications. To close this gap, we have created a deep learning model for road extraction (hereafter referred to as segmentation) in SAR and use it to compare standard model outputs against the aforementioned most popular methods for uncertainty estimation, MCD and DE. We demonstrate that these methods are not effective as an indicator of segmentation quality when measuring uncertainty (as indicated by model softmax outputs) across an entire image but are effective when uncertainty is measured from the set of road predictions only. Furthermore, we show a marked improvement in the correlation between prediction uncertainty and segmentation quality when we increase the set of road predictions by including predictions with lower softmax scores. We demonstrate the efficacy of our application of MCD and DE methods with an experimental design that measures performance in real-world quality assessment using in-distribution (ID) and out-of-distribution (OOD) data. These results inform the development of mission-critical deep learning systems in remote sensing. Tasks in medical image analysis that have a similar morphology to road structures, such as blood vessel segmentation, can also benefit from our findings.

Automatic Modulation Classification (AMC) is of paramount importance in wireless communication systems. Existing methods usually adopt a single category of neural network or stack different categories of networks in series, and rarely extract different types of features simultaneously in a proper way. When it comes to the output layer, softmax function is applied for classification to expand the inter-class distance. In this paper, we propose a hybrid parallel network for the AMC problem. Our proposed method designs a hybrid parallel structure which utilizes Convolution Neural Network (CNN) and Gate Rate Unit (GRU) to extract spatial features and temporal features respectively. Instead of superposing these two categories of features directly, three different attention mechanisms are applied to assign weights for different types of features. Finally, a cosine similarity metric named Additive Margin softmax function, which can expand the inter-class distance and compress the intra-class distance simultaneously, is adopted for output. Simulation results demonstrate that the proposed method can achieve remarkable performance on an open access dataset.