Vector Quantization Research Articles

A dynamic programming-optimized two-layer adaptive energy management strategy for electric vehicles considering driving pattern recognition

To achieve optimal real-time power allocation in electric vehicles, a two-layer adaptive dynamic programming (DP) optimization energy management strategy (EMS) has been proposed. The upper layer uses learning vector quantization (LVQ) to produce real-time driving pattern recognition (DPR) results. The method determines 10 characteristic parameters for training the recognition network and the length of the sampling window is 120 s. The typical driving cycles are divided into different levels of blocks to identify the real-time DPR level. The lower layer adopts the optimization strategy of DP extraction to adjust the power distribution between the battery pack and the supercapacitor pack according to the recognition results. DP is used to define a cost function to minimize the energy loss of the hybrid energy storage system (HESS) and optimize the battery usage range in the system. The near-optimal real-time EMS is extracted by analyzing the DP control behavior of the battery under the layered state of charge (SOC). The simulation results indicate that the proposed new rule control based on DP optimization (NRB) EMS improves the system efficiency by 10 % compared with the original rule-based (RB) EMS under different temperatures and DPR levels. In addition, the system efficiency gap is controlled at 3 % compared with DP.

An Efficient Fuzzy based MRI Brain Tumor Segmentation and Classification

Picture division is a significant testing factor in clinical picture division. This paper portrays the division strategy comprising two stages. In the initial step, the MRI cerebrum picture is gained from the patients' information base, in that film, relics and commotion are taken out after the Binarization strategy is applied for picture division. The Binaraization works with the assistance of the Fuzzy C Means Clustering calculation, hence the calculation assumes the principle part in the framework, in this most minimal level of the weight vector, higher worth of cancer pixels, calculation speed is accomplished by the Fuzzy C Mean with vector quantization. The point of this exploration work is to give an assortment of fluffy c-implies (FCM) calculation that gives picture bunching utilizing the MRI Brain Tumor data set. The proposed calculation joins the neighborhood spatial data and dark level data in an original fluffy manner. The new calculation is called Fuzziness Confined Message C-Means (FCM2). FCM2 can defeat the drawbacks of the known fluffy c-implies calculations and simultaneously improves the grouping execution. The significant trait of FCM2 is the utilization of a fluffy neighborhood (both spatial and dark level) closeness measure, expecting to ensure clamor heartlessness and picture detail conservation. Besides, the proposed calculation Experiments performed on manufactured and true pictures show that the FCM2 calculation is powerful and proficient, giving strength to uproarious pictures.

#6868 MACHINE LEARNING-BASED HYPERTENSION DISEASE RISK CLASSIFICATION USING LEARNING VECTOR QUANTIZATION ALGORITHM

Abstract Background and Aims According to the World Health Organization, the disease dubbed as the silent killer of death is the number one cause of death in the world. Hypertension is a disease that does not show certain symptoms. There are about 95% of hypertension cases have no known cause, while the rest are caused by other diseases. Therefore, an early detection system for hypertension is needed. In this research, the Learning Vector Quantization (LVQ) algorithm is used to perform the classification process. Method In this research, the LVQ machine learning algorithm is used. LVQ is an algorithm that is a type of artificial neural network and uses neural computation. The data in this study is hypertension patient data from Sultan Daeng Radja Hospital which is divided into four classes. In the system design stage using the LVQ algorithm, the process in this system is by entering input data in the form of medical records consisting of 12 attributes based on their weights which will then be carried out by the LVQ process to produce classification results. If the classification results have been obtained, the system will stop. After the classification testing process is carried out, the accuracy calculation process is carried out. Results Based on six test scenarios that have been carried out, the system produces recommendations for a value of 0.1 for the learning rate, a value of 0.2 for the learning rate multiplier, the amount of training data of as much as 50%, the maximum epoch value of six, the minimum value of alpha 0.001. From these values, the average accuracy result obtained is 94%. Conclusion Based on the value of accuracy results obtained exceeding 90%, the system can be done quite successfully in detecting hypertension. However, further research is needed to further develop the system.

Combination of Paper and Electronic Trail Making Tests for Automatic Analysis of Cognitive Impairment: Development and Validation Study.

Computer-aided detection, used in the screening and diagnosing of cognitive impairment, provides an objective, valid, and convenient assessment. Particularly, digital sensor technology is a promising detection method. This study aimed to develop and validate a novel Trail Making Test (TMT) using a combination of paper and electronic devices. This study included community-dwelling older adult individuals (n=297), who were classified into (1) cognitively healthy controls (HC; n=100 participants), (2) participants diagnosed with mild cognitive impairment (MCI; n=98 participants), and (3) participants with Alzheimer disease (AD; n=99 participants). An electromagnetic tablet was used to record each participant's hand-drawn stroke. A sheet of A4 paper was placed on top of the tablet to maintain the traditional interaction style for participants who were not familiar or comfortable with electronic devices (such as touchscreens). In this way, all participants were instructed to perform the TMT-square and circle. Furthermore, we developed an efficient and interpretable cognitive impairment-screening model to automatically analyze cognitive impairment levels that were dependent on demographic characteristics and time-, pressure-, jerk-, and template-related features. Among these features, novel template-based features were based on a vector quantization algorithm. First, the model identified a candidate trajectory as the standard answer (template) from the HC group. The distance between the recorded trajectories and reference was computed as an important evaluation index. To verify the effectiveness of our method, we compared the performance of a well-trained machine learning model using the extracted evaluation index with conventional demographic characteristics and time-related features. The well-trained model was validated using follow-up data (HC group: n=38; MCI group: n=32; and AD group: n=22). We compared 5 candidate machine learning methods and selected random forest as the ideal model with the best performance (accuracy: 0.726 for HC vs MCI, 0.929 for HC vs AD, and 0.815 for AD vs MCI). Meanwhile, the well-trained classifier achieved better performance than the conventional assessment method, with high stability and accuracy of the follow-up data. The study demonstrated that a model combining both paper and electronic TMTs increases the accuracy of evaluating participants' cognitive impairment compared to conventional paper-based feature assessment.

Vector Quantized Semantic Communication System

Although analog semantic communication systems have received considerable attention in the literature, there is less work on digital semantic communication systems. In this paper, we develop a deep learning (DL)-enabled vector quantized (VQ) semantic communication system for image transmission, named VQ-DeepSC. Specifically, we propose a convolutional neural network (CNN)-based transceiver to extract multi-scale semantic features of images and introduce multi-scale semantic embedding spaces to perform semantic feature quantization, rendering the data compatible with digital communication systems. Furthermore, we employ adversarial training to improve the quality of received images by introducing a PatchGAN discriminator. Experimental results demonstrate that the proposed VQ-DeepSC is more robustness than BPG in digital communication systems and has comparable MS-SSIM performance to the DeepJSCC method.

Leaf Disease Detection in Android

Abstract: In agriculture, early disease identification is crucial for a productive crop product. Ails similar as bacterial spot, late scar, Septoria splint spot, and unheroic twisted splint the quality of the tomato crop. Automatic bracket ways of factory conditions also help in taking action once they're discovered diseased splint symptoms Presented below is a Convolutional Learning Vector Quantization and Neural Network( CNN) model system for detecting tomato splint dis- ease grounded on the( LVQ) algorithm and categorization. There are 500 tomato prints in the dataset. leaves that display four complaint symptoms. We created a model of CNN for point birth and categorization automatically. Color Research on factory splint conditions laboriously uses information. In our model, three channels grounded on RGB are subordinated to pollutants

Leaf Disease Detection

Abstract: Early disease identification is crucial for productive crop production in agriculture. illnesses such as bacterial spot, late blight, Septoria leaf spot, and yellow curved leaf the quality of the tomato harvest. Automatic classification techniques of plant diseases also assist in taking action once they are discovered diseased leaf symptoms Presented below is a Convolutional Learning Vector Quantization and Neural Network (CNN) model Method for detecting tomato leaf disease based on the (LVQ) algorithm and categorization. There are 500 tomato photos in the dataset. leaves that display four disease symptoms. We created a model of CNN for feature extraction and categorization automatically. Color Research on plant leaf diseases actively uses information. In our model, three channels based on RGB are subjected to filters.

Performance Augmentation of Cuckoo Search Optimization Technique Using Vector Quantization in Image Compression

For constructing the best local codebook for image compression, there are many Vector Quantization (VQ) procedures, but the simplest VQ procedure is the Linde–Buzo–Gray (LBG) procedure. Techniques such as the Gaussian Dissemination Function (GDF) are used for the searching process in generating a global codebook for particle swarm optimization (PSO), Honeybee mating optimization (HBMO), and Firefly (FA) procedures. However, when particle velocity is very high, FA encounters a problem when brighter fireflies are trivial, and PSO suffers uncertainty in merging. A novel procedure, Cuckoo Search–Kekre Fast Codebook Generation (CS-KFCG), is proposed that enhances Cuckoo Search–Linde–Buzo–Gray (CS-LBG) codebook by implementing a Flight Dissemination Function (FDF), which produces more speed than other states of the art algorithms with appropriate mutation expectations for the overall codebook. Also, CS-KFGC has generated a high Peak Signal Noise Ratio (PSNR) in terms of high duration (time) and better acceptability rate.

Artificial Neural Networks for Diagnosis of Kidney Stones Disease

The foundation of medical care for children with renal stone disease is the assessment of metabolic risk factors, which aims to stop the growth of preexisting calculi and subsequent stone occurrences. In this retrospective analysis, 90 children with kidney stone disease who had been sent to our institution and had undergone clinical testing in accordance with a defined procedure had their metabolic risk factors, clinical histories, and family histories assessed. Our pediatric patients were 10.7 years old on average, with a male to female ratio of 1.14:1.0. In 84.4% of the instances, biochemical abnormalities were discovered. Of the patients, 52.2% (n = 47) had only one urine metabolic risk factor, whereas the remaining 31.1% (n = 28) had several risk factors. Adrenal hypercalciuria. The aim of this work is to compare the performance of all three neural networks on the basis of its accuracy, time taken to build model, and training data set size. We will use Learning vector quantization (LVQ), two layers feed forward perceptron trained with back propagation training algorithm and Radial basis function (RBF) networks for diagnosis of kidney stone disease. In this work we used Waikato Environment for Knowledge Analysis (WEKA) version 3.7.5 as simulation tool which is an open source tool

An efficient memory reserving-and-fading strategy for vector quantization based 3D brain segmentation and tumor extraction using an unsupervised deep learning network.

Deep learning networks are state-of-the-art approaches for 3D brain image segmentation, and the radiological characteristics extracted from tumors are of great significance for clinical diagnosis, treatment planning, and treatment outcome evaluation. However, two problems have been the hindering factors in brain image segmentation techniques. One is that deep learning networks require large amounts of manually annotated data. Another issue is the computational efficiency of 3D deep learning networks. In this study, we propose a vector quantization (VQ)-based 3D segmentation method that employs a novel unsupervised 3D deep embedding clustering (3D-DEC) network and an efficiency memory reserving-and-fading strategy. The VQ-based 3D-DEC network is trained on volume data in an unsupervised manner to avoid manual data annotation. The memory reserving-and-fading strategy beefs up model efficiency greatly. The designed methodology makes deep learning-based model feasible for biomedical image segmentation. The experiment is divided into two parts. First, we extensively evaluate the effectiveness and robustness of the proposed model on two authoritative MRI brain tumor databases (i.e., IBSR and BrainWeb). Second, we validate the model using real 3D brain tumor data collected from our institute for clinical practice significance. Results show that our method (without data manual annotation) has superior accuracy ( Tanimoto coefficient on IBSR, 97.5% TP and 97.7% TN on BrainWeb, and 91% Dice, 88% sensitivity and 87% specificity on real brain data) and remarkable efficiency (speedup ratio is 18.72 on IBSR, 31.16 on BrainWeb, 31.00 on real brain data) compared to the state-of-the-art methods. The results show that our proposed model can address the lacks of manual annotations, and greatly increase computation speedup with competitive segmentation accuracy compared to other state-of-the-art 3D CNN models. Moreover, the proposed model can be used for tumor treatment follow-ups every 6months, providing critical details for surgical and postoperative treatment by correctly extracting numerical radiomic features of tumors.

A novel coal-rock cutting state identification model based on the Internet of Things

The sudden change in the interface between coal-rock mass can lead to the increased abrasion of picks and the failure rate of mining machinery. The safe and efficient coal-rock identification technology is the key to realize the intelligent control of coal mining machinery. To realize the real-time perception and accurate recognition of coal-rock cutting state information, a novel coal-rock cutting state identification model based on the Internet of Things (IoT) is explored. Specifically, by using the virtual prototype technology, multi-source heterogeneous data from acquisition, processing and identification of coal-rock cutting state information are fused and analyzed. First, the paper analyzes the physical and mechanical characteristics of coal-rock mass, and takes the cutting drum of bolter miner as the research object to theoretically analyze its load, which provides a foundation for the research on coal-rock cutting state identification. Second, a rigid-flexible coupling virtual prototype model of the cutting drum and coal-rock models under different cutting conditions are established. The simulation process is implemented by employing the discrete element method (DEM) to ensure the real-time transmission of motion information and state characteristic signals. Ultimately, the dynamic information of drum load during the coal-rock cutting process is obtained. Finally, LVQ (Learning Vector Quantization) and PSO-BP (Particle Swarm Optimization-Back Propagation) neural networks are created, and the variation coefficient, waveform factor, and peak factor of load curves of cutting drums in the coal-rock mass with different firmness coefficients are input into the neural networks as feature vectors for state recognition. The experimental results show that LVQ and PSO-BP neural networks can be used for coal-rock cutting state identification, and PSO-BP network has faster convergence speed and higher recognition efficiency, which provides a new scheme for coal-rock cutting state identification to improve the safety and efficiency of coal mining machinery.

Cross Attention Transformers for Multi-modal Unsupervised Whole-Body PET Anomaly Detection

Cancer is a highly heterogeneous condition that can occur almost anywhere in the human body. [<sup>18</sup>F]fluorodeoxyglucose Positron Emission Tomography (<sup>18</sup>F-FDG PET) is a imaging modality commonly used to detect cancer due to its high sensitivity and clear visualisation of the pattern of metabolic activity. Nonetheless, as cancer is highly heterogeneous, it is challenging to train general-purpose discriminative cancer detection models, with data availability and disease complexity often cited as a limiting factor. Unsupervised learning methods, more specifically anomaly detection models, have been suggested as a putative solution. These models learn a healthy representation of tissue and detect cancer by predicting deviations from the healthy norm, which requires models capable of accurately learning long-range interactions between organs, their imaging patterns, and other abstract features with high levels of expressivity. Such characteristics are suitably satisfied by transformers, which have been shown to generate state-of-the-art results in unsupervised anomaly detection by training on normal data. This work expands upon such approaches by introducing multi-modal conditioning of the transformer via cross-attention i.e. supplying anatomical reference information from paired CT images to aid the PET anomaly detection task. Furthermore, we show the importance and impact of codebook sizing within a Vector Quantized Variational Autoencoder, on the ability of the transformer network to fulfill the task of anomaly detection. Using 294 whole-body PET/CT samples containing various cancer types, we show that our anomaly detection method is robust and capable of achieving accurate cancer localization results even in cases where normal training data is unavailable. In addition, we show the efficacy of this approach on out-of-sample data showcasing the generalizability of this approach even with limited training data. Lastly, we propose to combine model uncertainty with a new kernel density estimation approach, and show that it provides clinically and statistically significant improvements in accuracy and robustness, when compared to the classic residual-based anomaly maps. Overall, a superior performance is demonstrated against leading state-of-the-art alternatives, drawing attention to the potential of these approaches.

Speech to Text Conversion System for Myanmar Alphabet

This paper is aimed to implement the speech to text conversion system for Myanmar alphabet. The Myanmar alphabet consists of 33 characters from ?ka? to ?ah?. The proposed system is software architecture which allows the user to speak against the computer in Myanmar language and the corresponding character is printed on the screen in the Microsoft Office Word Document Format. The system is emphasized on Speaker Independent Isolated Word Recognition System. The proposed system directly acquires and converts speech to text. This system contains two main modules: feature extraction and feature matching. Mel Frequency Cepstrum Coefficients (MFCC) is applied for feature extraction which extracts a small amount of data from the voice signal that can later be used to represent each character. Feature matching involves the actual procedure to identify the unknown character by comparing extracted features from the voice inputs of a set of known characters. In this system, Vector Quantization (VQ) approach using Linde, Buzo and Gray (LBG) clustering algorithm, which reduces the amount of data and complexity, is applied for feature matching. To implement this system MATLAB programming language is used.

Compression of images with a mathematical approach based on sine and cosine equations and vector quantization (VQ)

Compression of images with a mathematical approach based on sine and cosine equations and vector quantization (VQ)

A Physical Layer Key Generation Scheme Based on Deep Learning Compensation and Balanced Vector Quantization

Channel reciprocity is the foundation for physical layer key generation, which is influenced by noise, hardware impairments, and synchronization offsets. Weak channel reciprocity will result in a high key disagreement rate (KDR). The existing solutions for improving channel reciprocity cannot achieve satisfactory performance improvements. Furthermore, the existing quantization algorithms generally use one-dimensional channel features to quantize and generate secret keys, which cannot fully utilize channel information. The multidimensional vector quantization technique also needs to improve in terms of randomness and time complexity. This paper proposes a physical layer key generation scheme based on deep learning and balanced vector quantization. Specifically, we build a channel reciprocity compensation network (CRCNet) to learn the mapping relationship between Alice and Bob’s channel measurements. Alice compensates for channel measurements via a trained CRCNet to reduce channel measurement errors between legitimate users and enhance channel reciprocity. We also propose a balanced vector quantization algorithm based on integer linear programming (ILP-BVQ). ILP-BVQ reduces the time complexity of quantization on the basis of ensuring key randomness and a low KDR. Simulation results showed that the proposed CRCNet performs better in terms of channel reciprocity and KDR, while the proposed ILP-BVQ algorithm improves time consumption and key randomness.

Introduction of k-means clustering into random cascade model for disaggregation of rainfall from daily to 1-hour resolution with improved preservation of extreme rainfall

Temporal disaggregation of rainfall has been of particular focus because of the non-availability of higher-resolution rainfall data for a long-duration period. Fine temporal resolution rainfall is used in a multitude of hydrological applications. Researchers have proposed various disaggregation models to disaggregate coarse temporal resolution rainfall. In this paper, firstly, the microcanonical multiplicative random cascade (MMRC) model is applied for disaggregation from daily rainfall to a one-hour scale. The model is applied in four different rainfall stations for disaggregation having varying rainfall patterns and characteristics. It is observed that the MMRC model can generate statistically reliable rainfall time series; however, the extreme rainfall characteristics are not well conserved by the model for all the stations.This paper describes a new model based on a random multiplicative cascade process where classification and parameter generation is done by k-means clustering such that it can better conserve extreme rainfall conditions and generate a reliable rainfall time series (MMRC-K). K-means clustering is a vector quantization method that divides the observations into a particular number of clusters based on the nearest mean called cluster centroid. The novel approach is tested with the same four Indian cities. The use of k-means clustering has made the classification and parameter generation of the model robust such that it can work with data sets of varying characteristics. It is found that MMRC-K provides improved conservation of extreme rainfall characteristics compared to the MMRC model for all four stations. The MMRC-K model reproduces the IDF curves of Delhi and Mumbai stations quite well; however, a little discrepancy was observed at higher resolution and larger return periods in Kolkata and Chennai stations. Extreme rainfall at finer resolution is used in various hydrological analyses and design problems like urban drainage design, stormwater management, etc. The overall superior conservation of the extreme rainfall characteristics in the model-generated rainfall time series by the MMRC-K model compared to the MMRC model supports the potential applicability of the model for temporal disaggregation.

A Novel Image Compression Algorithm Based on Autoencoder for Biometric Template Protection Scheme

Objectives: To evaluate the proposed deep learning-based method for biometric template protection schemes with other existing image compression methods. Methods: In this study, four image compression methods such as JPEG_LS, Vector quantization (VQ), Run-length encoding (RLE), and the Autoencoder methods are implemented. The experimental results are compared with the different performance parameters that are applied over the CASIA I iris dataset possessing 756 images. Findings: The proposed autoencoder method algorithm offered enhanced results in terms of PSNR, SSIM, MSE, and CR when compared to other transform methods. The deep neural network-based autoencoder algorithm achieved the highest compression ratio of 89.05 percent, while the conventional algorithms achieved the highest image quality rate of 94.05 percent. Novelty: A novel autoencoder-based image compression model has been proposed in this study. The proposed Autoencoder (AE) method incorporates five stages like Initialization, Learning compression representations, Generation of chaotic sequence, Image encryption and decryption, and Image reconstruction. The generation of the chaotic sequence using the stochastic logistic map, as well as the learning compression representations contributes towards the novelty of this study. Keywords: Image compression; Autoencoder; Runlength encoding (RLE); Vector quantization

Electroencephalography (EEG) eye state classification using learning vector quantization and bagged trees

The analysis of Electroencephalography (EEG) signals has been an effective way of eye state identification. Its significance is highlighted by studies that examined the classification of eye states using machine learning techniques. In previous studies, supervised learning techniques have been widely used in EEG signals analysis for eye state classification. Their main goal has been the improvement of classification accuracy through the use of novel algorithms. The trade-off between classification accuracy and computation complexity is an important task in EEG signals analysis. In this paper, a hybrid method that can handle multivariate signals and non-linear is proposed with supervised and un-supervised learning to achieve a fast EEG eye state classification with high prediction accuracy to provide real-time decision-making applicability. We use the Learning Vector Quantization (LVQ) technique and bagged tree techniques. The method was evaluated on a real-world EEG dataset which included 14976 instances after the removal of outlier instances. Using LVQ, 8 clusters were generated from the data. The bagged tree was applied on 8 clusters and compared with other classifiers. Our experiments revealed that LVQ combined with the bagged tree provides the best results (Accuracy = 0.9431) compared with the bagged tree, CART (Classification And Regression Tree) (Accuracy = 0.8200), LDA (Linear Discriminant Analysis) (Accuracy = 0.7931), Random Trees (Accuracy = 0.8311), Naïve Bayes (Accuracy = 0.8331) and Multilayer Perceptron (Accuracy = 0.7718), which demonstrates the effectiveness of incorporating ensemble learning and clustering approaches in the analysis of EEG signals. We also provided the time complexity of the methods for prediction speed (Observation/Second). The result showed that LVQ + Bagged Tree provides the best result for prediction speed (58942 Obs/Sec) in relation to Bagged Tree (28453 Obs/Sec), CART (27784 Obs/Sec), LDA (26435 Obs/Sec), Random Trees (27921), Naïve Bayes (27217) and Multilayer Perceptron (24163).

Learning Vector Quantization, Hebbian Learning, and Self-Organizing Map for Classification

Deep Learning has been rapidly developed. Almost all proposed methods already have very high accuracy. Most of these methods still use techniques from the past with some modifications to adapt to existing modules. Sometimes it is necessary to understand past methods to produce new methods. Therefore, this research examines past models that have the potential to improve the performance of existing deep learning models. The methods to be examined include Learning Vector Quantization (LVQ), Hebbian learning, and Self-Organizing Map (SOM). The iris dataset available on Scikit-learn (SKlearn) is used here for testing in cases of supervised learning and unsupervised learning (especially SOM). The results show that LVQ has a good accuracy of 93%, while Hebbian learning has an accuracy of 56%. SOM fluctuates between 88% and 93%. Although the accuracy of SOM does not exceed LVQ, this model does not require labels in its training process.

Quantum Computing Approaches for Vector Quantization-Current Perspectives and Developments.

In the field of machine learning, vector quantization is a category of low-complexity approaches that are nonetheless powerful for data representation and clustering or classification tasks. Vector quantization is based on the idea of representing a data or a class distribution using a small set of prototypes, and hence, it belongs to interpretable models in machine learning. Further, the low complexity of vector quantizers makes them interesting for the application of quantum concepts for their implementation. This is especially true for current and upcoming generations of quantum devices, which only allow the execution of simple and restricted algorithms. Motivated by different adaptation and optimization paradigms for vector quantizers, we provide an overview of respective existing quantum algorithms and routines to realize vector quantization concepts, maybe only partially, on quantum devices. Thus, the reader can infer the current state-of-the-art when considering quantum computing approaches for vector quantization.