scholarly journals Object detection and object classification using machine learning Algorithms

2020 ◽  
Vol 2 (3) ◽  
pp. 21-29
Author(s):  
Dora Racheed ◽  
Rahmatullah Muin ◽  
Ahmed Jaylan
Keyword(s):  
Computational Cost ◽  
Machine Learning Algorithms ◽  
Learning Approaches ◽  
Feature Maps ◽  
Improve Performance ◽  
Wide Margin ◽  
Satisfactory Performance ◽  
Computational Reduction ◽  
Variable Representation ◽  
Conventional Detection

Urban objects are characterized by a very variable representation in terms of shape, texture and color. In addition, they are present multiple times on the images to be analyzed and can be stuck to each other. To carry out the automatic localization and recognition of the different objects we propose to use supervised learning approaches. Due to their characteristics, urban objects are difficult to detect and conventional detection approaches do not offer satisfactory performance. We proposed the use of a wide margin separator network (SVM) in order to better merge the information from the different resolutions and therefore to improve the representativeness of the urban object. The use of an SVM network makes it possible to improve performance but at a significant computational cost. We then proposed to use an activation path making it possible to reduce complexity without losing efficiency. This path will activate the network sequentially and stop the exploration when the probability of detecting an object is high. In the case of a location based on the extraction of characteristics then the classification, the computational reduction is a factor of five. Subsequently, we have shown that we can combine the SVM network with feature maps from convolutional neural networks.

scholarly journals Random Shifting for CNN: a Solution to Reduce Information Loss in Down-Sampling Layers

2017 ◽  
Author(s):  
Gangming Zhao ◽  
Jingdong Wang ◽  
Zhaoxiang Zhang
Keyword(s):  
Neural Networks ◽  
Computational Cost ◽  
Receptive Fields ◽  
Information Loss ◽  
Network Architectures ◽  
Training Process ◽  
Feature Maps ◽  
Improve Performance ◽  
Deep Convolutional Neural Networks ◽  
Random Strategy

Down-sampling is widely adopted in deep convolutional neural networks (DCNN) for reducing the number of network parameters while preserving the transformation invariance. However, it cannot utilize information effectively because it only adopts a fixed stride strategy, which may result in poor generalization ability and information loss. In this paper, we propose a novel random strategy to alleviate these problems by embedding random shifting in the down-sampling layers during the training process. Random shifting can be universally applied to diverse DCNN models to dynamically adjust receptive fields by shifting kernel centers on feature maps in different directions. Thus, it can generate more robust features in networks and further enhance the transformation invariance of down-sampling operators. In addition, random shifting cannot only be integrated in all down-sampling layers including strided convolutional layers and pooling layers, but also improve performance of DCNN with negligible additional computational cost. We evaluate our method in different tasks (e.g., image classification and segmentation) with various network architectures (i.e., AlexNet, FCN and DFN-MR). Experimental results demonstrate the effectiveness of our proposed method.

Prediction of K562 Cells Functional Inhibitors Based on Machine Learning Approaches

2020 ◽  
Vol 25 (40) ◽  
pp. 4296-4302 ◽  
Author(s):  
Yuan Zhang ◽  
Zhenyan Han ◽  
Qian Gao ◽  
Xiaoyi Bai ◽  
Chi Zhang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Inclusion Bodies ◽  
Cross Validation ◽  
Independent Set ◽  
K562 Cells ◽  
Machine Learning Algorithms ◽  
Learning Approaches ◽  
Validation Test ◽  
Excess Number ◽  
Fold Cross Validation

Background: β thalassemia is a common monogenic genetic disease that is very harmful to human health. The disease arises is due to the deletion of or defects in β-globin, which reduces synthesis of the β-globin chain, resulting in a relatively excess number of α-chains. The formation of inclusion bodies deposited on the cell membrane causes a decrease in the ability of red blood cells to deform and a group of hereditary haemolytic diseases caused by massive destruction in the spleen. Methods: In this work, machine learning algorithms were employed to build a prediction model for inhibitors against K562 based on 117 inhibitors and 190 non-inhibitors. Results: The overall accuracy (ACC) of a 10-fold cross-validation test and an independent set test using Adaboost were 83.1% and 78.0%, respectively, surpassing Bayes Net, Random Forest, Random Tree, C4.5, SVM, KNN and Bagging. Conclusion: This study indicated that Adaboost could be applied to build a learning model in the prediction of inhibitors against K526 cells.

scholarly journals Towards an Efficient CNN Inference Architecture Enabling In-Sensor Processing

Sensors ◽  
2021 ◽  
Vol 21 (6) ◽  
pp. 1955
Author(s):  
Md Jubaer Hossain Pantho ◽  
Pankaj Bhowmik ◽  
Christophe Bobda
Keyword(s):  
Power Consumption ◽  
High Speed ◽  
Image Sensor ◽  
Machine Learning Algorithms ◽  
Hierarchical Optimization ◽  
Computation Method ◽  
Optimization Approach ◽  
Efficient Computation ◽  
Feature Maps ◽  
Dynamic Power

The astounding development of optical sensing imaging technology, coupled with the impressive improvements in machine learning algorithms, has increased our ability to understand and extract information from scenic events. In most cases, Convolution neural networks (CNNs) are largely adopted to infer knowledge due to their surprising success in automation, surveillance, and many other application domains. However, the convolution operations’ overwhelming computation demand has somewhat limited their use in remote sensing edge devices. In these platforms, real-time processing remains a challenging task due to the tight constraints on resources and power. Here, the transfer and processing of non-relevant image pixels act as a bottleneck on the entire system. It is possible to overcome this bottleneck by exploiting the high bandwidth available at the sensor interface by designing a CNN inference architecture near the sensor. This paper presents an attention-based pixel processing architecture to facilitate the CNN inference near the image sensor. We propose an efficient computation method to reduce the dynamic power by decreasing the overall computation of the convolution operations. The proposed method reduces redundancies by using a hierarchical optimization approach. The approach minimizes power consumption for convolution operations by exploiting the Spatio-temporal redundancies found in the incoming feature maps and performs computations only on selected regions based on their relevance score. The proposed design addresses problems related to the mapping of computations onto an array of processing elements (PEs) and introduces a suitable network structure for communication. The PEs are highly optimized to provide low latency and power for CNN applications. While designing the model, we exploit the concepts of biological vision systems to reduce computation and energy. We prototype the model in a Virtex UltraScale+ FPGA and implement it in Application Specific Integrated Circuit (ASIC) using the TSMC 90nm technology library. The results suggest that the proposed architecture significantly reduces dynamic power consumption and achieves high-speed up surpassing existing embedded processors’ computational capabilities.

scholarly journals Searching for improvements in predicting human eye colour from DNA

2021 ◽  
Author(s):  
Magdalena Kukla-Bartoszek ◽  
Paweł Teisseyre ◽  
Ewelina Pośpiech ◽  
Joanna Karłowska-Pik ◽  
Piotr Zieliński ◽  
...  
Keyword(s):  
Machine Learning ◽  
Regression Models ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Sequencing Analysis ◽  
Learning Approaches ◽  
Human Eye ◽  
Software Analysis ◽  
Whole Exome ◽  
Eye Colour

AbstractIncreasing understanding of human genome variability allows for better use of the predictive potential of DNA. An obvious direct application is the prediction of the physical phenotypes. Significant success has been achieved, especially in predicting pigmentation characteristics, but the inference of some phenotypes is still challenging. In search of further improvements in predicting human eye colour, we conducted whole-exome (enriched in regulome) sequencing of 150 Polish samples to discover new markers. For this, we adopted quantitative characterization of eye colour phenotypes using high-resolution photographic images of the iris in combination with DIAT software analysis. An independent set of 849 samples was used for subsequent predictive modelling. Newly identified candidates and 114 additional literature-based selected SNPs, previously associated with pigmentation, and advanced machine learning algorithms were used. Whole-exome sequencing analysis found 27 previously unreported candidate SNP markers for eye colour. The highest overall prediction accuracies were achieved with LASSO-regularized and BIC-based selected regression models. A new candidate variant, rs2253104, located in the ARFIP2 gene and identified with the HyperLasso method, revealed predictive potential and was included in the best-performing regression models. Advanced machine learning approaches showed a significant increase in sensitivity of intermediate eye colour prediction (up to 39%) compared to 0% obtained for the original IrisPlex model. We identified a new potential predictor of eye colour and evaluated several widely used advanced machine learning algorithms in predictive analysis of this trait. Our results provide useful hints for developing future predictive models for eye colour in forensic and anthropological studies.

scholarly journals A Joint 2D-3D Complementary Network for Stereo Matching

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1430
Author(s):  
Xiaogang Jia ◽  
Wei Chen ◽  
Zhengfa Liang ◽  
Xin Luo ◽  
Mingfei Wu ◽  
...  
Keyword(s):  
Stereo Matching ◽  
Computational Cost ◽  
Research Field ◽  
Disparity Map ◽  
Improve Performance ◽  
Cost Aggregation ◽  
Disparity Range ◽  
Public Datasets ◽  
Coarse To Fine ◽  
Speed And Accuracy

Stereo matching is an important research field of computer vision. Due to the dimension of cost aggregation, current neural network-based stereo methods are difficult to trade-off speed and accuracy. To this end, we integrate fast 2D stereo methods with accurate 3D networks to improve performance and reduce running time. We leverage a 2D encoder-decoder network to generate a rough disparity map and construct a disparity range to guide the 3D aggregation network, which can significantly improve the accuracy and reduce the computational cost. We use a stacked hourglass structure to refine the disparity from coarse to fine. We evaluated our method on three public datasets. According to the KITTI official website results, Our network can generate an accurate result in 80 ms on a modern GPU. Compared to other 2D stereo networks (AANet, DeepPruner, FADNet, etc.), our network has a big improvement in accuracy. Meanwhile, it is significantly faster than other 3D stereo networks (5× than PSMNet, 7.5× than CSN and 22.5× than GANet, etc.), demonstrating the effectiveness of our method.

Emotion recognition using time–frequency ridges of EEG signals based on multivariate synchrosqueezing transform

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Ahmet Mert ◽  
Hasan Huseyin Celik
Keyword(s):  
Emotion Recognition ◽  
Feature Vector ◽  
Dimensional Space ◽  
Computational Cost ◽  
Feature Space ◽  
Maximum Energy ◽  
Machine Learning Algorithms ◽  
Eeg Signals ◽  
Time Frequency ◽  
Synchrosqueezing Transform

Abstract The feasibility of using time–frequency (TF) ridges estimation is investigated on multi-channel electroencephalogram (EEG) signals for emotional recognition. Without decreasing accuracy rate of the valence/arousal recognition, the informative component extraction with low computational cost will be examined using multivariate ridge estimation. The advanced TF representation technique called multivariate synchrosqueezing transform (MSST) is used to obtain well-localized components of multi-channel EEG signals. Maximum-energy components in the 2D TF distribution are determined using TF-ridges estimation to extract instantaneous frequency and instantaneous amplitude, respectively. The statistical values of the estimated ridges are used as a feature vector to the inputs of machine learning algorithms. Thus, component information in multi-channel EEG signals can be captured and compressed into low dimensional space for emotion recognition. Mean and variance values of the five maximum-energy ridges in the MSST based TF distribution are adopted as feature vector. Properties of five TF-ridges in frequency and energy plane (e.g., mean frequency, frequency deviation, mean energy, and energy deviation over time) are computed to obtain 20-dimensional feature space. The proposed method is performed on the DEAP emotional EEG recordings for benchmarking, and the recognition rates are yielded up to 71.55, and 70.02% for high/low arousal, and high/low valence, respectively.

Lightweight deep residual network for alzheimer’s disease classification using sMRI slices

2021 ◽  
pp. 1-9
Author(s):  
Yanteng Zhang ◽  
Qizhi Teng ◽  
Linbo Qing ◽  
Yan Liu ◽  
Xiaohai He
Keyword(s):  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Deep Learning ◽  
Computational Cost ◽  
Disease Classification ◽  
Learning Networks ◽  
Feature Maps ◽  
Medical Field ◽  
Widespread Application ◽  
Healthy Control

Alzheimer’s disease (AD) is a degenerative brain disease and the most common cause of dementia. In recent years, with the widespread application of artificial intelligence in the medical field, various deep learning-based methods have been applied for AD detection using sMRI images. Many of these networks achieved AD vs HC (Healthy Control) classification accuracy of up to 90%but with a large number of computational parameters and floating point operations (FLOPs). In this paper, we adopt a novel ghost module, which uses a series of cheap operations of linear transformation to generate more feature maps, embedded into our designed ResNet architecture for task of AD vs HC classification. According to experiments on the OASIS dataset, our lightweight network achieves an optimistic accuracy of 97.92%and its total parameters are dozens of times smaller than state-of-the-art deep learning networks. Our proposed AD classification network achieves better performance while the computational cost is reduced significantly.

scholarly journals Attack and Anomaly Detection in IoT Networks Using Supervised Machine Learning Approaches

2021 ◽  
Vol 35 (1) ◽  
pp. 11-21
Author(s):  
Himani Tyagi ◽  
Rajendra Kumar
Keyword(s):  
Machine Learning ◽  
Performance Metrics ◽  
Detection System ◽  
Feature Reduction ◽  
Machine Learning Algorithms ◽  
Supervised Machine Learning ◽  
Testing Time ◽  
Learning Approaches ◽  
Reduction Techniques ◽  
Share Data

IoT is characterized by communication between things (devices) that constantly share data, analyze, and make decisions while connected to the internet. This interconnected architecture is attracting cyber criminals to expose the IoT system to failure. Therefore, it becomes imperative to develop a system that can accurately and automatically detect anomalies and attacks occurring in IoT networks. Therefore, in this paper, an Intrsuion Detection System (IDS) based on extracted novel feature set synthesizing BoT-IoT dataset is developed that can swiftly, accurately and automatically differentiate benign and malicious traffic. Instead of using available feature reduction techniques like PCA that can change the core meaning of variables, a unique feature set consisting of only seven lightweight features is developed that is also IoT specific and attack traffic independent. Also, the results shown in the study demonstrates the effectiveness of fabricated seven features in detecting four wide variety of attacks namely DDoS, DoS, Reconnaissance, and Information Theft. Furthermore, this study also proves the applicability and efficiency of supervised machine learning algorithms (KNN, LR, SVM, MLP, DT, RF) in IoT security. The performance of the proposed system is validated using performance Metrics like accuracy, precision, recall, F-Score and ROC. Though the accuracy of Decision Tree (99.9%) and Randon Forest (99.9%) Classifiers are same but other metrics like training and testing time shows Random Forest comparatively better.

scholarly journals Analysis of Residual Current Flows in Inverter Based Energy Systems Using Machine Learning Approaches

Energies ◽  
2022 ◽  
Vol 15 (2) ◽  
pp. 582
Author(s):  
Holger Behrends ◽  
Dietmar Millinger ◽  
Werner Weihs-Sedivy ◽  
Anže Javornik ◽  
Gerold Roolfs ◽  
...  
Keyword(s):  
Machine Learning ◽  
Early Stage ◽  
Residual Current ◽  
Operating Conditions ◽  
Machine Learning Algorithms ◽  
Photovoltaic System ◽  
Detection Methods ◽  
Learning Approaches ◽  
Residual Currents ◽  
Current Flows

Faults and unintended conditions in grid-connected photovoltaic systems often cause a change of the residual current. This article describes a novel machine learning based approach to detecting anomalies in the residual current of a photovoltaic system. It can be used to detect faults or critical states at an early stage and extends conventional threshold-based detection methods. For this study, a power-hardware-in-the-loop approach was carried out, in which typical faults have been injected under ideal and realistic operating conditions. The investigation shows that faults in a photovoltaic converter system cause a unique behaviour of the residual current and fault patterns can be detected and identified by using pattern recognition and variational autoencoder machine learning algorithms. In this context, it was found that the residual current is not only affected by malfunctions of the system, but also by volatile external influences. One of the main challenges here is to separate the regular residual currents caused by the interferences from those caused by faults. Compared to conventional methods, which respond to absolute changes in residual current, the two machine learning models detect faults that do not affect the absolute value of the residual current.

Sign in / Sign up

Export Citation Format

Share Document