Multi-level feature fusion in densely connected deep-learning architecture and depth-first search for crack segmentation on images collected with smartphones

Cracks are important signs of degradation in existing infrastructure systems. Automatic crack detection and segmentation plays a key role in developing smart infrastructure systems. However, this field has been challenging over the last decades due to irregular shape of the cracks and complex illumination conditions. This article proposes a novel deep-learning architecture for crack segmentation at pixel-level. In this architecture, one convolutional layer is densely connected to multiple other layers in a feed-forward fashion. Max pooling layers are used to reduce the dimensions of the features, and transposed convolution layers are used for multi-level feature fusion. A depth-first search–based algorithm is applied as post-processing tool to remove isolated pixels and improve the accuracy. The method is tested on two previously published data sets. It can reach 92.02%, 91.13%, and 91.58% for the first data set, and 92.17%, 91.61%, and 91.89% for the second data set for precision, recall, and F1 score, respectively. The performance of the proposed method outperforms other state-of-the-art methods. At the end of the article, the influence of feature fusion methods and transfer learning are also discussed.

Download Full-text

Human Activity Recognition using Fourier Transform Inspired Deep Learning Combination Model

International Journal of Sensors Wireless Communications and Control ◽

10.2174/2210327908666180727123657 ◽

2019 ◽

Vol 9 (1) ◽

pp. 16-31

Author(s):

Kyungkoo Jun

Keyword(s):

Fourier Transform ◽

Deep Learning ◽

Short Term Memory ◽

Window Size ◽

Sensor Data ◽

Data Sets ◽

Data Set ◽

Proposed Model ◽

Testing Data ◽

Labeling Scheme

Background & Objective: This paper proposes a Fourier transform inspired method to classify human activities from time series sensor data. Methods: Our method begins by decomposing 1D input signal into 2D patterns, which is motivated by the Fourier conversion. The decomposition is helped by Long Short-Term Memory (LSTM) which captures the temporal dependency from the signal and then produces encoded sequences. The sequences, once arranged into the 2D array, can represent the fingerprints of the signals. The benefit of such transformation is that we can exploit the recent advances of the deep learning models for the image classification such as Convolutional Neural Network (CNN). Results: The proposed model, as a result, is the combination of LSTM and CNN. We evaluate the model over two data sets. For the first data set, which is more standardized than the other, our model outperforms previous works or at least equal. In the case of the second data set, we devise the schemes to generate training and testing data by changing the parameters of the window size, the sliding size, and the labeling scheme. Conclusion: The evaluation results show that the accuracy is over 95% for some cases. We also analyze the effect of the parameters on the performance.

Download Full-text

Investigating the impact of pre-processing techniques and pre-trained word embeddings in detecting Arabic health information on social media

Journal Of Big Data ◽

10.1186/s40537-021-00488-w ◽

2021 ◽

Vol 8 (1) ◽

Author(s):

Yahya Albalawi ◽

Jim Buckley ◽

Nikola S. Nikolov

Keyword(s):

Social Media ◽

Deep Learning ◽

Comprehensive Evaluation ◽

Classification Problem ◽

Data Sets ◽

Word Embeddings ◽

Data Set ◽

Lower Accuracy ◽

Health Related ◽

The Impact

AbstractThis paper presents a comprehensive evaluation of data pre-processing and word embedding techniques in the context of Arabic document classification in the domain of health-related communication on social media. We evaluate 26 text pre-processings applied to Arabic tweets within the process of training a classifier to identify health-related tweets. For this task we use the (traditional) machine learning classifiers KNN, SVM, Multinomial NB and Logistic Regression. Furthermore, we report experimental results with the deep learning architectures BLSTM and CNN for the same text classification problem. Since word embeddings are more typically used as the input layer in deep networks, in the deep learning experiments we evaluate several state-of-the-art pre-trained word embeddings with the same text pre-processing applied. To achieve these goals, we use two data sets: one for both training and testing, and another for testing the generality of our models only. Our results point to the conclusion that only four out of the 26 pre-processings improve the classification accuracy significantly. For the first data set of Arabic tweets, we found that Mazajak CBOW pre-trained word embeddings as the input to a BLSTM deep network led to the most accurate classifier with F1 score of 89.7%. For the second data set, Mazajak Skip-Gram pre-trained word embeddings as the input to BLSTM led to the most accurate model with F1 score of 75.2% and accuracy of 90.7% compared to F1 score of 90.8% achieved by Mazajak CBOW for the same architecture but with lower accuracy of 70.89%. Our results also show that the performance of the best of the traditional classifier we trained is comparable to the deep learning methods on the first dataset, but significantly worse on the second dataset.

Download Full-text

Relative testis size and mating systems in anurans: large testis in multiple-male mating in foam-nesting frogs

Animal Biology ◽

10.1163/157075511x570312 ◽

2011 ◽

Vol 61 (2) ◽

pp. 225-238 ◽

Cited By ~ 15

Author(s):

Wen Bo Liao ◽

Zhi Ping Mi ◽

Cai Quan Zhou ◽

Ling Jin ◽

Xian Han ◽

...

Keyword(s):

Sperm Competition ◽

Published Data ◽

Male Mating ◽

Data Sets ◽

Testis Size ◽

Data Set ◽

Monogamous Species ◽

Large Testis ◽

Testes Size ◽

Testis Mass

AbstractComparative studies of the relative testes size in animals show that promiscuous species have relatively larger testes than monogamous species. Sperm competition favours the evolution of larger ejaculates in many animals – they give bigger testes. In the view, we presented data on relative testis mass for 17 Chinese species including 3 polyandrous species. We analyzed relative testis mass within the Chinese data set and combining those data with published data sets on Japanese and African frogs. We found that polyandrous foam nesting species have relatively large testes, suggesting that sperm competition was an important factor affecting the evolution of relative testes size. For 4 polyandrous species testes mass is positively correlated with intensity (males/mating) but not with risk (frequency of polyandrous matings) of sperm competition.

Download Full-text

Children with 5′-end NF1 gene mutations are more likely to have glioma

Neurology Genetics ◽

10.1212/nxg.0000000000000192 ◽

2017 ◽

Vol 3 (5) ◽

pp. e192 ◽

Cited By ~ 12

Author(s):

Corina Anastasaki ◽

Stephanie M. Morris ◽

Feng Gao ◽

David H. Gutmann

Keyword(s):

Gene Mutation ◽

Statistical Significance ◽

Gene Mutations ◽

Neurofibromatosis Type ◽

Published Data ◽

Data Sets ◽

Nonsense Mutations ◽

Data Set ◽

Nf1 Gene ◽

The Relationship

Objective:To ascertain the relationship between the germline NF1 gene mutation and glioma development in patients with neurofibromatosis type 1 (NF1).Methods:The relationship between the type and location of the germline NF1 mutation and the presence of a glioma was analyzed in 37 participants with NF1 from one institution (Washington University School of Medicine [WUSM]) with a clinical diagnosis of NF1. Odds ratios (ORs) were calculated using both unadjusted and weighted analyses of this data set in combination with 4 previously published data sets.Results:While no statistical significance was observed between the location and type of the NF1 mutation and glioma in the WUSM cohort, power calculations revealed that a sample size of 307 participants would be required to determine the predictive value of the position or type of the NF1 gene mutation. Combining our data set with 4 previously published data sets (n = 310), children with glioma were found to be more likely to harbor 5′-end gene mutations (OR = 2; p = 0.006). Moreover, while not clinically predictive due to insufficient sensitivity and specificity, this association with glioma was stronger for participants with 5′-end truncating (OR = 2.32; p = 0.005) or 5′-end nonsense (OR = 3.93; p = 0.005) mutations relative to those without glioma.Conclusions:Individuals with NF1 and glioma are more likely to harbor nonsense mutations in the 5′ end of the NF1 gene, suggesting that the NF1 mutation may be one predictive factor for glioma in this at-risk population.

Download Full-text

Multimodal biometrics of fingerprint and signature recognition using multi-level feature fusion and deep learning techniques

Indonesian Journal of Electrical Engineering and Computer Science ◽

10.11591/ijeecs.v22.i1.pp187-195 ◽

2021 ◽

Vol 22 (1) ◽

pp. 187

Author(s):

Arjun Benagatte Channegowda ◽

H N Prakash

Keyword(s):

Deep Learning ◽

Feature Fusion ◽

Research Work ◽

Recognition Rate ◽

Recognition System ◽

Multimodal Biometrics ◽

Histograms Of Oriented Gradients ◽

Multi Level ◽

Deep Learning Neural Network ◽

Hidden Neurons

Providing security in biometrics is the major challenging task in the current situation. A lot of research work is going on in this area. Security can be more tightened by using complex security systems, like by using more than one biometric trait for recognition. In this paper multimodal biometric models are developed to improve the recognition rate of a person. The combination of physiological and behavioral biometrics characteristics is used in this work. Fingerprint and signature biometrics characteristics are used to develop a multimodal recognition system. Histograms of oriented gradients (HOG) features are extracted from biometric traits and for these feature fusions are applied at two levels. Features of fingerprint and signatures are fused using concatenation, sum, max, min, and product rule at multilevel stages, these features are used to train deep learning neural network model. In the proposed work, multi-level feature fusion for multimodal biometrics with a deep learning classifier is used and results are analyzed by a varying number of hidden neurons and hidden layers. Experiments are carried out on SDUMLA-HMT, machine learning and data mining lab, Shandong University fingerprint datasets, and MCYT signature biometric recognition group datasets, and encouraging results were obtained.

Download Full-text

Classification of Clinically Significant Prostate Cancer on Multi-Parametric MRI: A Validation Study Comparing Deep Learning and Radiomics

Cancers ◽

10.3390/cancers14010012 ◽

2021 ◽

Vol 14 (1) ◽

pp. 12

Author(s):

Jose M. Castillo T. ◽

Muhammad Arif ◽

Martijn P. A. Starmans ◽

Wiro J. Niessen ◽

Chris H. Bangma ◽

...

Keyword(s):

Prostate Cancer ◽

Deep Learning ◽

Characteristic Curve ◽

Model Development ◽

Learning Model ◽

Multiparametric Mri ◽

Data Sets ◽

Data Set ◽

Test Sets ◽

Deep Learning Model

The computer-aided analysis of prostate multiparametric MRI (mpMRI) could improve significant-prostate-cancer (PCa) detection. Various deep-learning- and radiomics-based methods for significant-PCa segmentation or classification have been reported in the literature. To be able to assess the generalizability of the performance of these methods, using various external data sets is crucial. While both deep-learning and radiomics approaches have been compared based on the same data set of one center, the comparison of the performances of both approaches on various data sets from different centers and different scanners is lacking. The goal of this study was to compare the performance of a deep-learning model with the performance of a radiomics model for the significant-PCa diagnosis of the cohorts of various patients. We included the data from two consecutive patient cohorts from our own center (n = 371 patients), and two external sets of which one was a publicly available patient cohort (n = 195 patients) and the other contained data from patients from two hospitals (n = 79 patients). Using multiparametric MRI (mpMRI), the radiologist tumor delineations and pathology reports were collected for all patients. During training, one of our patient cohorts (n = 271 patients) was used for both the deep-learning- and radiomics-model development, and the three remaining cohorts (n = 374 patients) were kept as unseen test sets. The performances of the models were assessed in terms of their area under the receiver-operating-characteristic curve (AUC). Whereas the internal cross-validation showed a higher AUC for the deep-learning approach, the radiomics model obtained AUCs of 0.88, 0.91 and 0.65 on the independent test sets compared to AUCs of 0.70, 0.73 and 0.44 for the deep-learning model. Our radiomics model that was based on delineated regions resulted in a more accurate tool for significant-PCa classification in the three unseen test sets when compared to a fully automated deep-learning model.

Download Full-text

Long-Term Rainfall Forecast Model Based on The TabNet and LightGbm Algorithm

10.21203/rs.3.rs-107107/v1 ◽

2020 ◽

Author(s):

Tianyu Xu ◽

Yongchuan Yu ◽

Jianzhuo Yan ◽

Hongxia Xu

Keyword(s):

Prediction Model ◽

Feature Fusion ◽

Forecast Model ◽

Data Sets ◽

Good Prediction ◽

Data Set ◽

Rainfall Prediction ◽

Improve Model ◽

Probability Prediction

Abstract Due to the problems of unbalanced data sets and distribution differences in long-term rainfall prediction, the current rainfall prediction model had poor generalization performance and could not achieve good prediction results in real scenarios. This study uses multiple atmospheric parameters (such as temperature, humidity, atmospheric pressure, etc.) to establish a TabNet-LightGbm rainfall probability prediction model. This research uses feature engineering (such as generating descriptive statistical features, feature fusion) to improve model accuracy, Borderline Smote algorithm to improve data set imbalance, and confrontation verification to improve distribution differences. The experiment uses 5 years of precipitation data from 26 stations in the Beijing-Tianjin-Hebei region of China to verify the proposed rainfall prediction model. The test set is to predict the rainfall of each station in one month. The experimental results shows that the model has good performance with AUC larger than 92%. The method proposed in this study further improves the accuracy of rainfall prediction, and provides a reference for data mining tasks.

Download Full-text

Recommendations for bacterial ribosome profiling experiments based on bioinformatic evaluation of published data

Journal of Biological Chemistry ◽

10.1074/jbc.ra119.012161 ◽

2020 ◽

Vol 295 (27) ◽

pp. 8999-9011 ◽

Cited By ~ 2

Author(s):

Alina Glaub ◽

Christopher Huptas ◽

Klaus Neuhaus ◽

Zachary Ardern

Keyword(s):

Ribosome Profiling ◽

Published Data ◽

Data Sets ◽

Drug Induced ◽

Data Set ◽

Protein Coding ◽

Bacterial Ribosome ◽

Translation Start ◽

Selection Step ◽

Basic Characteristics

Ribosome profiling (RIBO-Seq) has improved our understanding of bacterial translation, including finding many unannotated genes. However, protocols for RIBO-Seq and corresponding data analysis are not yet standardized. Here, we analyzed 48 RIBO-Seq samples from nine studies of Escherichia coli K12 grown in lysogeny broth medium and particularly focused on the size-selection step. We show that for conventional expression analysis, a size range between 22 and 30 nucleotides is sufficient to obtain protein-coding fragments, which has the advantage of removing many unwanted rRNA and tRNA reads. More specific analyses may require longer reads and a corresponding improvement in rRNA/tRNA depletion. There is no consensus about the appropriate sequencing depth for RIBO-Seq experiments in prokaryotes, and studies vary significantly in total read number. Our analysis suggests that 20 million reads that are not mapping to rRNA/tRNA are required for global detection of translated annotated genes. We also highlight the influence of drug-induced ribosome stalling, which causes bias at translation start sites. The resulting accumulation of reads at the start site may be especially useful for detecting weakly expressed genes. As different methods suit different questions, it may not be possible to produce a “one-size-fits-all” ribosome profiling data set. Therefore, experiments should be carefully designed in light of the scientific questions of interest. We propose some basic characteristics that should be reported with any new RIBO-Seq data sets. Careful attention to the factors discussed should improve prokaryotic gene detection and the comparability of ribosome profiling data sets.

Download Full-text

MFVT:An Anomaly Traffic Detection Method Merging Feature Fusion Network and Vision Transformer Architecture

10.21203/rs.3.rs-877144/v1 ◽

2021 ◽

Author(s):

Ming Li ◽

Dezhi Han ◽

Dun Li ◽

Han Liu ◽

Chin- Chen Chang

Keyword(s):

Deep Learning ◽

Intrusion Detection ◽

Network Traffic ◽

Feature Fusion ◽

Imbalanced Data ◽

Network Intrusion Detection ◽

Detection Accuracy ◽

Data Sets ◽

Imbalanced Data Sets ◽

Network Intrusion

Abstract Network intrusion detection, which takes the extraction and analysis of network traffic features as the main method, plays a vital role in network security protection. The current network traffic feature extraction and analysis for network intrusion detection mostly uses deep learning algorithms. Currently, deep learning requires a lot of training resources, and have weak processing capabilities for imbalanced data sets. In this paper, a deep learning model (MFVT) based on feature fusion network and Vision Transformer architecture is proposed, to which improves the processing ability of imbalanced data sets and reduces the sample data resources needed for training. Besides, to improve the traditional raw traffic features extraction methods, a new raw traffic features extraction method (CRP) is proposed, the CPR uses PCA algorithm to reduce all the processed digital traffic features to the specified dimension. On the IDS 2017 dataset and the IDS 2012 dataset, the ablation experiments show that the performance of the proposed MFVT model is significantly better than other network intrusion detection models, and the detection accuracy can reach the state-of-the-art level. And, When MFVT model is combined with CRP algorithm, the detection accuracy is further improved to 99.99%.

Download Full-text

Research on Network Security Application Based on Deep Learning

CONVERTER ◽

10.17762/converter.235 ◽

2021 ◽

pp. 598-605

Author(s):

Zhao Jianchao

Keyword(s):

Deep Learning ◽

Learning Algorithm ◽

Rapid Development ◽

Internet Security ◽

Detection Accuracy ◽

Data Sets ◽

Learning Technology ◽

Data Set ◽

Deep Learning Algorithm ◽

Internet Industry

Behind the rapid development of the Internet industry, Internet security has become a hidden danger. In recent years, the outstanding performance of deep learning in classification and behavior prediction based on massive data makes people begin to study how to use deep learning technology. Therefore, this paper attempts to apply deep learning to intrusion detection to learn and classify network attacks. Aiming at the nsl-kdd data set, this paper first uses the traditional classification methods and several different deep learning algorithms for learning classification. This paper deeply analyzes the correlation among data sets, algorithm characteristics and experimental classification results, and finds out the deep learning algorithm which is relatively good at. Then, a normalized coding algorithm is proposed. The experimental results show that the algorithm can improve the detection accuracy and reduce the false alarm rate.

Download Full-text