Vector Quantization Research Articles

Adaptive Modem Based on LSTM-AutoEncoder with Vector Quantization

Recently, researchers have achieved the goal of using unified architecture to achieve multiple modulation modes under specific conditions. However, existing research still suffers from the problem of large resource and time overhead. This paper proposes an adaptive modem based on a vector quantization (VQ) long short-term memory autoencoder (LSTM-AE), designed to implement modulation and demodulation of signals from the second to thirty-second order in a lightweight way. By leveraging the memory capacity of the LSTM module and the compression capability of the autoencoder, the model is able to support multi-order modulation methods. This study used the Adam optimizer for training and testing on a simple dataset extended by adding AWGN noise only and made modifications based on the MSE loss function to balance the accuracy and training speed of each part of the model. Experiments demonstrate that the proposed method not only achieves comparable modem performance to existing frameworks for second to thirty-second order signals, but also significantly reduces the number of parameters and training time. Experimental results indicate that the proposed methodology not only matches the performance of existing frameworks for signals ranging from the second to the thirty-second order, but also employs merely 79.6% of the average parameter count and a mere 7.4% of the average training duration. This represents a substantial reduction in resource expenditure.

Intelligent Image Compression Model on the Basis of Wavelet Transform and Optimized Fuzzy C-Means-Based Vector Quantisation

Intelligent Image Compression Model on the Basis of Wavelet Transform and Optimized Fuzzy C-Means-Based Vector Quantisation

Towards Robust Colour Texture Analysis with Limited Training Data

Texture analysis plays an important role in different domains of healthcare, agriculture, and industry, where multi-channel sensors are gaining more attention. This contribution presents an interpretable and efficient framework for texture classification and segmentation that exploits colour or channel information and does not require much data to produce accurate results. This makes such a framework well-suited for medical applications and resource-limited hardware. Our approach builds upon a distance-based generalized matrix learning vector quantization (GMLVQ) algorithm. We extend it with parametrized angle-based dissimilarity and introduce a special matrix format for multi-channel images. Classification accuracy evaluation of various model designs was performed on VisTex and ALOT data, and the segmentation application was demonstrated on an agricultural data set. Our extension of parametrized angle dissimilarity measure leads to better model generalization and robustness against varying lighting conditions than its Euclidean counterpart. The proposed matrix format for multichannel images enhances classification accuracy while reducing the number of parameters. Regarding segmentation, our method shows promising results, provided with a small class-imbalanced training data set. Proposed methodology achieves higher accuracy than prior work benchmarks and a small-scale CNN while maintaining a significantly lower parameter count. Notably, it is interpretable and accurate in scenarios where limited and unbalanced training data are available.

Speaker Recognition: A Comparative analysis Between Deep Learning and Non-Deep Learning Methodologies

This work carries out a comparative study of two methodologies for speaker recognition. It is Deep Learning (DL) and Vector Quantization (VQ). The key area in biometric au-thentication systems involves speaker recognition. This requires durable and efficient algorithms. The aim is to ensure a high accuracy and reliability. The study delves into a deep neural network (DNN) model implementation. It leverages advanced feature extraction. It uses pattern recognition capabilities which are in DL. The study also examines a traditional VQ approach. This method makes use of codebook generation and quantization for speaker ID. Extensive experimentation was done on standard datasets. The project evaluates the performance of two methods. It compares accuracy. It assesses computational complexity. It does so for noise and for variations in speech. The findings of this analysis reveal the strengths and limitations of each technique. Looking at their practical applicability in real-world scenarios provides insights. The comparative results of these techniques aim to guide future developments. This concerns speaker recognition systems - particularly their potential for enhanced performance.

SIFG: an ensemble model for sieving fake news from genuine without metadata by combining syntactic and semantic features

ABSTRACT With the emergence of the Internet, social media, and smartphones, almost everyone can now claim that they have news in their pockets. This phenomenon makes people aware of their surroundings and knowledgeable of current events. But, it also presents new challenges, such as automatic detection, machine/bot-generated fake news, limited or no metadata, etc. In this paper, we mitigate some of these challenges by taking a step toward SIeving Fake News from Genuine (SIFG) and presenting a system named SIFG. In SIFG we have used basic and advanced Natural Language Processing techniques to detect online fake news automatically. This makes SIFG independent of the metadata, such as the source, network structure and behavior, temporal propagation, and responses, about the news. We also introduce specific Parts of Speech patterns, utilize Generalized Relevance Learning Vector Quantization for feature selection, and employ ensemble learning to improve the performance of SIFG. When tested with a publicly available online fake news dataset of 10,000 fake and 10,000 genuine news, SIFG shows promising results and achieves an accuracy in the range of 89.85% − 95%.

Progressive updates of convolutional neural networks for enhanced reliability in small satellite applications

Small satellites enable many important applications for both economic and scientific purposes. Many of these applications are inherently data-centric and rely on large amounts of high-resolution satellite imagery to be delivered in a timely manner. However, communicating this data to Earth is challenging due to intermittent connectivity, high packet losses, low data rates, and similar issues. Therefore, efficient onboard prioritization and data processing are essential for future satellite missions. Machine learning methods, such as deep neural networks, are very suitable for such prioritization, as they are already used extensively for satellite imagery processing and they can be deployed onboard of satellites. However, updating them to support new classification requirements when the satellite is already in orbit is difficult, as often multiple passes are required to complete model transmission due to the communication challenges. To cope with this issue, we propose a progressive transmission mechanism for model updates, which leverages vector quantization and arithmetic coding. Our mechanism allows to achieve high accuracies even with partially updated models. Evaluation results show that our mechanism significantly outperforms other less optimized transmission schemes.

Predicting Number of Vehicles Involved in Rural Crashes Using Learning Vector Quantization Algorithm

Roads represent very important infrastructure and play a significant role in economic, cultural, and social growth. Therefore, there is a critical need for many researchers to model crash injury severity in order to study how safe roads are. When measuring the cost of crashes, the severity of the crash is a critical criterion, and it is classified into various categories. The number of vehicles involved in the crash (NVIC) is a crucial factor in all of these categories. For this purpose, this research examines road safety and provides a prediction model for the number of vehicles involved in a crash. Specifically, learning vector quantization (LVQ 2.1), one of the sub-branches of artificial neural networks (ANNs), is used to build a classification model. The novelty of this study demonstrates LVQ 2.1’s efficacy in categorizing accident data and its ability to improve road safety strategies. The LVQ 2.1 algorithm is particularly suitable for classification tasks and works by adjusting prototype vectors to improve the classification performance. The research emphasizes how urgently better prediction algorithms are needed to handle issues related to road safety. In this study, a dataset of 564 crash records from rural roads in Calabria between 2017 and 2048, a region in southern Italy, was utilized. The study analyzed several key parameters, including daylight, the crash type, day of the week, location, speed limit, average speed, and annual average daily traffic, as input variables to predict the number of vehicles involved in rural crashes. The findings revealed that the “crash type” parameter had the most significant impact, whereas “location” had the least significant impact on the occurrence of rural crashes in the investigated areas.

Undersampling based on generalized learning vector quantization and natural nearest neighbors for imbalanced data

Undersampling based on generalized learning vector quantization and natural nearest neighbors for imbalanced data

Web-based Application for Diagnosis of Diabetes using Learning Vector Quantization (LVQ)

Diabetes is a chronic disease that causes the most deaths in the world. This disease can cause long-term complications that develop gradually, such as heart attacks, strokes, and problems with the kidneys, eyes, skin, and blood vessels. Therefore, early diagnosis of diabetes is crucial for patients to know their diabetes status. In this study, we designed a web-based application for diabetes diagnosis using Learning Vector Quantization (LVQ). The dataset was collected from Kaggle's Diabetes Dataset which contains eight attributes, namely pregnancy, glucose, blood pressure, insulin, skin thickness, BMI, diabetes lineage function, and age, with two classes, namely negative diabetes (healthy) and positive diabetes. The results show that the best accuracy is 73.1% with a learning rate of 0.001. These findings can help patients detect diabetes problems early.

Vector Quantization With Error Uniformly Distributed Over an Arbitrary Set

Vector Quantization With Error Uniformly Distributed Over an Arbitrary Set

Generating extremely low-dimensional representation of subsurface earth models using vector quantization and deep Autoencoder

Geological model compression is crucial for making large and complex models more manageable. By reducing the size of these models, compression techniques enable efficient storage, enhance computational efficiency, making it feasible to perform complex simulations and analyses in a shorter time. This is particularly important in applications such as reservoir management, groundwater hydrology, and geological carbon storage, where large geomodels with millions of grid cells are common. This study presents a comprehensive overview of previous work on geomodel compression and introduces several autoencoder-based deep-learning architectures for low-dimensional representation of modified Brugge-field geomodels. The compression and reconstruction efficiencies of autoencoders (AE), variational autoencoders (VAE), vector-quantized variational autoencoders (VQ-VAE), and vector-quantized variational autoencoders 2 (VQ-VAE2) were tested and compared to the traditional singular value decomposition (SVD) method. Results show that the deep-learning-based approaches significantly outperform SVD, achieving higher compression ratios while maintaining or even exceeding the reconstruction quality. Notably, VQ-VAE2 achieves the highest compression ratio of 667:1 with a structural similarity index metric (SSIM) of 0.92, far surpassing the 10:1 compression ratio of SVD with a SSIM of 0.9. The result of this work shows that, unlike traditional approaches, which often rely on linear transformations and can struggle to capture complex, non-linear relationships within geological data, VQ-VAE's use of vector quantization helps in preserving high-resolution details and enhances the model's ability to generalize across varying geological complexities.

Determination of early breeder in goldfish (Carassius auratus Linn.) with learning vector quantization, probabilistic and pattern recognition neural networks

It is common practice to categorize growing fish according to quality in fish farming centers. For such an application, experienced and estimating people are needed. By pre-selecting the right fish for the ornamental fish industry, it gets ahead of the competitors. This study deals with the classification of early breeder determination in goldfish using three different Artificial Neural Network (ANN) techniques. This classification model can help the fish industry assess classification risks and make the right decision. The used dataset was derived from the results of the classification section of 120 goldfish. It consisted of 7 input parameters (day, live weight, body length, head height, head width, body height, and current class). During trial, all goldfish fed by a diet contained 360 g crude protein and 4449.85 kcal metabolizable energy (kg / dry matter). The important types of classification ANNs, namely Learning Vector Quantization Neural Network (LVQNN), Probabilistic Neural Network (PNN), and Pattern Recognition Neural Network (PRNN) were employed for the machine learning scheme. The training and test performances of the ANN models were compared with the correct prediction ratio. They showed that all of the proposed ANN techniques were well at classification. However, the PRNN model was better than the LVQNN and PNN as a classifier for the breeder selection of goldfish. Therefore, the results of PRNN model such as histogram, Receiver Operating Characteristic (ROC) curve, regression, confusion matrix, accuracy, sensitivity, precision, and F1 score were given and discussed. Furthermore, the classification of early breeder determination in goldfish were examined for days with the best PRNN.

Imitation Learning from a Single Demonstration Leveraging Vector Quantization for Robotic Harvesting

The ability of robots to tackle complex non-repetitive tasks will be key in bringing a new level of automation in agricultural applications still involving labor-intensive, menial, and physically demanding activities due to high cognitive requirements. Harvesting is one such example as it requires a combination of motions which can generally be broken down into a visual servoing and a manipulation phase, with the latter often being straightforward to pre-program. In this work, we focus on the task of fresh mushroom harvesting which is still conducted manually by human pickers due to its high complexity. A key challenge is to enable harvesting with low-cost hardware and mechanical systems, such as soft grippers which present additional challenges compared to their rigid counterparts. We devise an Imitation Learning model pipeline utilizing Vector Quantization to learn quantized embeddings directly from visual inputs. We test this approach in a realistic environment designed based on recordings of human experts harvesting real mushrooms. Our models can control a cartesian robot with a soft, pneumatically actuated gripper to successfully replicate the mushroom outrooting sequence. We achieve 100% success in picking mushrooms among distractors with less than 20 min of data collection comprising a single expert demonstration and auxiliary, non-expert, trajectories. The entire model pipeline requires less than 40 min of training on a single A4000 GPU and approx. 20 ms for inference on a standard laptop GPU.

A geometric approach to robust medical image segmentation

Robustness of deep learning segmentation models is crucial for their safe incorporation into clinical practice. However, these models can falter when faced with distributional changes. This challenge is evident in magnetic resonance imaging (MRI) scans due to the diverse acquisition protocols across various domains, leading to differences in image characteristics such as textural appearances. We posit that the restricted anatomical differences between subjects could be harnessed to refine the latent space into a set of shape components. The learned set then aims to encompass the relevant anatomical shape variation found within the patient population.We explore this by utilising multiple MRI sequences to learn texture invariant and shape equivariant features which are used to construct a shape dictionary using vector quantisation. We investigate shape equivariance to a number of different types of groups. We hypothesise and prove that the greater the group order, i.e., the denser the constraint, the better becomes the model robustness. We achieve shape equivariance either with a contrastive based approach or by imposing equivariant constraints on the convolutional kernels. The resulting shape equivariant dictionary is then sampled to compose the segmentation output. Our method achieves state-of-the-art performance for the task of single domain generalisation for prostate and cardiac MRI segmentation. Code is available at https://github.com/AinkaranSanthi/A_Geometric_Perspective_For_Robust_Segmentation.

Patch-wise vector quantization for unsupervised medical anomaly detection

Radiography images inherently possess globally consistent structures while exhibiting significant diversity in local anatomical regions, making it challenging to model their normal features through unsupervised anomaly detection. Since unsupervised anomaly detection methods localize anomalies by utilizing discrepancies between learned normal features and input abnormal features, previous studies introduce a memory structure to capture the normal features of radiography images. However, these approaches store extremely localized image segments in their memory, causing the model to represent both normal and pathological features with the stored components. This poses a significant challenge in unsupervised anomaly detection by reducing the disparity between learned features and abnormal features. Furthermore, with the diverse settings in radiography imaging, the above issue is exacerbated: more diversity in the normal images results in stronger representation of pathological features. To resolve the issues above, we propose a novel pathology detection method called Patch-wise Vector Quantization (P-VQ). Unlike the previous methods, P-VQ learns vector-quantized representations of normal “patches” while preserving its spatial information by incorporating vector similarity metric. Furthermore, we introduce a novel method for selecting features in the memory to further enhance the robustness against diverse imaging settings. P-VQ even mitigates the “index collapse” problem of vector quantization by proposing top-k% dropout. Our extensive experiments on the BMAD benchmark demonstrate the superior performance of P-VQ against existing state-of-the-art methods.

Intelligent support in manufacturing process selection based on artificial neural networks, fuzzy logic, and genetic algorithms: Current state and future perspectives

Technological advances, dynamic customer needs, growing uncertainty, and the imperative for sustainable development pressure manufacturing entities to enhance productivity and competitiveness. In this challenging landscape, decision-making in manufacturing process selection is critical. Adopting intelligent support is essential for balancing performance and costs through optimal process selection. Through a comprehensive review of 93 studies published between 2013 and 2023, this paper aims to provide a profound understanding of intelligent support in manufacturing process selection. The findings, which indicate significant interest in intelligent methodologies for manufacturing process selection, are of great importance. Fuzzy logic is prevalent in additive manufacturing due to its ability to handle complex and imprecise data. At the same time, artificial neural networks are favored in conventional manufacturing for leveraging extensive historical data. Genetic algorithms are primarily used for optimization challenges. As manufacturing evolves with new technologies and complex materials, this paper advocates adopting a generalized matrix learning vector quantization neural network for efficient and intelligent process selection in additive and conventional approaches due to its capacity to leverage historical data and handle complex and high dimensional data.

MEASUREMENT OF CLASSIFICATION PERFORMANCE WITH THE LEARNING VECTOR QUANTIZATION METHOD ON COVID-19 VACCINATION DATA AT THE PARUMPANAI HEALTH CENTER

In the midst of the COVID-19 pandemic, various countries are always trying their best to restore global stability. One effective way is the discovery of several vaccines to prevent transmission of the virus. Indonesia is one of the countries that is aggressively implementing the COVID-19 vaccination. The vaccination process which has been carried out from February 2021 until the end of 2021 has covered approximately 160 million people or 76.83% of the target set by the government. Vaccine recipients have criteria to be able to get vaccinated to avoid side effects or complications. So it is necessary to classify groups that can receive vaccines and also delay vaccination. This research aims to determine the performance of the learning vector quantization classification method. Learning vector quantization method classification produces 95% accuracy, 97% precision, and 96% sensitivity. From these performance measurements, it can be concluded that the learning vector quantization method is very good and can be used in the classification of COVID-19 vaccination recipients at the Parumpanai Public Health Center, East Luwu Regency.

P‐57: A Novel Compression Algorithm using Machine Learning for Mura Compensation of OLED Panel

In the modern display panel manufacturing industry, the Mura problem that occurs during initial shipment has a significant impact on improving image quality. The purpose of this paper is to effectively solve these Mura defects while reducing additional production costs. We propose an innovative method of storing and compressing the De‐Mura compensation data of the display panel in external memory. Two‐dimensional compensation data compression based on machine learning using surface fitting and vector quantization is presented as an effective response to the Mura problem. This approach, which achieves up to 8:1 compression ratio while minimizing image quality degradation through vector quantization, is expected to provide economic benefits to manufacturers. The paper also introduces a new surface prediction method that replaces the existing DC‐shift and uses residual vector quantization, which can reduce the implementation complexity of vector quantization. Through experimental results, this study demonstrates that the compensated data compression technique yields excellent results in post‐compression restoration performance.

Vector Quantization Methods for Access Point Placement in Cell-Free Massive MIMO Systems

Vector Quantization Methods for Access Point Placement in Cell-Free Massive MIMO Systems

Fast Linde-Buzo-Gray (FLBG) Algorithm for Image Compression through Rescaling Using Bilinear Interpolation.

Vector quantization (VQ) is a block coding method that is famous for its high compression ratio and simple encoder and decoder implementation. Linde-Buzo-Gray (LBG) is a renowned technique for VQ that uses a clustering-based approach for finding the optimum codebook. Numerous algorithms, such as Particle Swarm Optimization (PSO), the Cuckoo search algorithm (CS), bat algorithm, and firefly algorithm (FA), are used for codebook design. These algorithms are primarily focused on improving the image quality in terms of the PSNR and SSIM but use exhaustive searching to find the optimum codebook, which causes the computational time to be very high. In our study, our algorithm enhances LBG by minimizing the computational complexity by reducing the total number of comparisons among the codebook and training vectors using a match function. The input image is taken as a training vector at the encoder side, which is initialized with the random selection of the vectors from the input image. Rescaling using bilinear interpolation through the nearest neighborhood method is performed to reduce the comparison of the codebook with the training vector. The compressed image is first downsized by the encoder, which is then upscaled at the decoder side during decompression. Based on the results, it is demonstrated that the proposed method reduces the computational complexity by 50.2% compared to LBG and above 97% compared to the other LBG-based algorithms. Moreover, a 20% reduction in the memory size is also obtained, with no significant loss in the image quality compared to the LBG algorithm.