Accurate modulation classification under impaired wireless channels via shallow convolutional neural networks

Charts are often used for the graphical representation of tabular data. Due to their vast expansion in various fields, it is necessary to develop computer algorithms that can easily retrieve and process information from chart images in a helpful way. Convolutional neural networks (CNNs) have succeeded in various image processing and classification tasks. Nevertheless, the success of training neural networks in terms of result accuracy and computational requirements requires careful construction of the network layers’ and networks’ parameters. We propose a novel Shallow Convolutional Neural Network (SCNN) architecture for chart-type classification and image generation. We validate the proposed novel network by using it in three different models. The first use case is a traditional SCNN classifier where the model achieves average classification accuracy of 97.14%. The second use case consists of two previously introduced SCNN-based models in parallel, with the same configuration, shared weights, and parameters mirrored and updated in both models. The model achieves average classification accuracy of 100%. The third proposed use case consists of two distinct models, a generator and a discriminator, which are both trained simultaneously using an adversarial process. The generated chart images are plausible to the originals. Extensive experimental analysis end evaluation is provided for the classification task of seven chart classes. The results show that the proposed SCNN is a powerful tool for chart image classification and generation, comparable with Deep Convolutional Neural Networks (DCNNs) but with higher efficiency, reduced computational time, and space complexity.

Automatic eyeglasses detection plays a major role in many facial analysis systems. To improve the robustness of these systems and cope with real-world applications, a high-speed eyeglasses detector that can achieve high accuracy is needed. Recent studies indicate that the features extracted from convolutional neural networks are compelling. Therefore, this paper presents an effective and efficient method for eyeglasses detection in facial images based on extracting deep features from a well-designed shallow convolutional neural network (CNN). The main contribution of this paper is to address the two essential aspects of CNN: (1) the size of the training dataset required and (2) the depth of the network architecture. To this end, we initialize the learning parameters of the shallow CNN by the parameters of a deep CNN which is fine-tuned on a small dataset. The depth of the neural network is then decreased by removing some convolutional layers after testing its performance on the validation dataset. As a result, a significantly more accurate shallow CNN architecture, Shallow-GlassNet, is obtained, which achieves not only high accuracy but also high speed in eyeglasses detection. Evaluation experiments have been conducted on two large unconstrained facial image databases, LFW and Celeb Faces. The results have demonstrated the superior performance of the proposed framework which achieves a mean accuracy of 99.73%.

Manifold causes of image blurring make the no-reference evaluation of realistic blurred images very challenging. Previous studies indicate that handcrafted features suffer from poor representation of the intrinsic characteristics of image blurring and thus blind image sharpness assessment (BISA) is unsatisfactory. This paper explores a shallow convolutional neural network (CNN) to address this problem facilitated by data augmentation. Superior to algorithms that necessitates considerable expertise and efforts to handcraft features for optimal representation of perceptual image quality, CNN directly integrates the retrieval of intrinsic features and the prediction of image blur quality into an optimization process. Moreover, experiments on Realistic Blurring Image Database have verified that CNN advances in retrieving intrinsic features and obtains good correlation with subjective image blurring evaluations.

Automatic modulation classification (AMC) plays an important role in many fields to identify the modulation type of wireless signals. In this paper, we introduce deep learning to signal recognition. Based on architecture analysis of the convolutional neural network (CNN), we used real signal data generated by instruments as dataset, and proposed an improved CNN architecture to achieve compatible recognition accuracy of modulation classification. According to various conditions of signal noise ratio (SNR), we test the proposed CNN architecture with the real sampled signals. Experiments results show that the high-layer network is not necessary for modulation recognition with high SNR signals. The proposed CNN architecture has higher average classification accuracy than RESNET and is more compatible for modulation classification of signals with lower SNR.

A fingerprint is an impression left by the friction ridges of a human finger. A fingerprint classification system groups fingerprint according to their characteristics and therefore helps to match a fingerprint against an extensive database of fingerprints. The Henry classification system is widely used among fingerprint classification systems. Some researchers have used traditional machine learning or deep learning for fingerprint classification. Nevertheless, traditional algorithms cannot extract the depth features of the fingerprint, and most deep learning algorithms lack fingerprint image enhancement. So, this paper combined the Gabor Filter and Convolutional Neural Network to extract fingerprint features. The model has two channels, one is a Deep Convolutional Neural Network (DCNN), and the other is a Shallow Convolutional Neural Network (SCNN). The DCNN consists of a neural network with eight layers, which can extract deep features of the fingerprint. The SCNN consists of Gabor Filter and a neural network with two layers that can extract features from clear fingerprint images. This paper uses NIST Special Database 4 for experiments. Experimental results show that the model proposed in this paper has achieved 91.4% accuracy. Compared with other algorithms, this model has higher accuracy than others. It shows that combined with the Gabor Filter and Convolutional Neural Network can better extract the ridge features of fingerprint images.

Automatic modulation classification receives significant interest in the context of current and future wireless communication systems. Deep learning emerged as a powerful tool for modulation classification, as it allows for joint discriminative features learning and signal classification. However, the optimization of deep neural network architectures for modulation classification is a manual and time-consuming process that requires profound domain knowledge and much effort. Most state-of-the-art solutions focus mainly on classification accuracy, while optimization of network complexity is neglected. This paper presents a novel bi-objective memetic algorithm, BO-NSMA, to search optimal deep neural network architectures for modulation classification to maximize classification accuracy and minimize network complexity. The experiments show that BO-NSMA, with an initial population of six individuals and only ten generations, finds a deep neural network architecture that outperforms all human-crafted architectures. Furthermore, BO-NSMA discovered the first low-complexity Convolutional neural network architecture, which achieves slightly better performance than costly Recurrent neural network architectures, allowing a 2.9-fold reduction in network complexity with 1.43% performance improvement. Compared to counterparts from network architecture search, BO-NSMA finds the best architecture, which achieves up to 18.73% accuracy gain and up to an 82-fold reduction in network complexity. The results are validated using the Wilcoxon signed-rank test.

Shallow convolutional neural networks (CNNs) have successfully been used to classify polarimetric synthetic aperture radar (PolSAR) imagery. However, one drawback of the existing deep CNN-based techniques is that the input PolSAR training data are often insufficient due to their need for a significant number of training data compared to shallow CNN models utilized in PolSAR image classification. In this paper, we propose using Haar wavelet transform in deep CNNs for effective feature extraction to improve the classification accuracy of PolSAR imagery. Based on the results, the proposed deep CNN model obtained better average accuracy in the San Francisco region with an accuracy of 93.3% and produced more homogeneous classification maps with less noise compared to the two much shallower CNN models of AlexNet (87.8%) and a 2D CNN network (91%). The proposed algorithm is efficient and may be applied over large areas to support regional wetland mapping and monitoring activities using PolSAR imagery. The codes are available at (https://github.com/aj1365/DeepCNN_Polsar).

Modulation classification, an intermediate step between signal detection and demodulation, is commonly deployed in many modern wireless communication systems. Although many approaches have been introduced in the last decades for identifying the modulation format of the incoming signal, they have the obstacle of mining radio characteristics for most traditional machine learning algorithms. To effectively handle this limitation, we propose an accurate modulation classification method by exploiting deep learning for being compatible with constellation diagram. A convolutional neural network (CNN), namely CRNet, is developed to proficiently learn the most relevant radio characteristics from transformed gray-scale constellation image by cross-residual connection, a novel structure for associating the intrinsic information between two processing flows specified by regular and grouped convolutional layers. Based on the experimental evaluation, CRNet achieves the classification rate of approximately 90% at +10 dB signal-to-noise ratio (SNR) under a multipath Rayleigh fading channel and further performs more accurately than some existing deep models for constellation-based modulation classification.

Detection and classification of radars based on pulses they transmit is an\nimportant application in electronic warfare systems. In this work, we propose a\nnovel deep-learning based technique that automatically recognizes intra-pulse\nmodulation types of radar signals. Re-assigned spectrogram of measured radar\nsignal and detected outliers of its instantaneous phases filtered by a special\nfunction are used for training multiple convolutional neural networks.\nAutomatically extracted features from the networks are fused to distinguish\nfrequency and phase modulated signals. Simulation results show that the\nproposed FF-CNN (Feature Fusion based Convolutional Neural Network) technique\noutperforms the current state-of-the-art alternatives and is easily scalable\namong broad range of modulation types.\n

Context. The article considers the features of solving non-traditional problems of recognition of specified types modulation signals in automated radio monitoring. The practical features of this problem determine the increased a priori uncertainty, which consists in the absence of a priori information about the distribution densities of the given signals and the presence of unknown signals. Objective. It is proposed to solve the problem using an unconventional method for the recognition of statistically specified random signals in the presence of a class of unknown signals. This method assumes that for the given signals there is a classified training sample of realizations, according to which the unknown parameters of their distributions are estimated, as well as some threshold values that determine the probabilities of correct recognition of the given types of signal modulation in the presence of unknown signals. Method. A general solution to the problem of recognition of given signals in the presence of unknown signals is given, and recognition methods of types modulation based on the description of signals by probabilistic model in the form of a mixture of distributions are given. The method is based on the description of signals by a probabilistic model in the form of a mixture of distributions and construction of a closed area for given signals in the probabilistic space of signals. Results. Studies of the recognition problems of given types of modulation of signals have been carried out. The studies were performed by statistical tests on samples of signals for radio monitoring of communications. In this case, the decisive rule for recognizing the given types of signal modulation is implemented in software on a computer. As a result of the statistical tests carried out on control samples of signals, estimates of the probabilities of correct recognition of the given types of signal modulation in the presence of unknown signals were obtained. Conclusions. Values of indicators of quality of radio emissions recognition acceptable for the practice of radio monitoring are obtained. The dependences of quality indicators on some conditions and recognition parameters are property. As a result of the research, practical recommendations were obtained on the use of the proposed method for recognizing specified types of signal modulation in automated radio monitoring systems.

Convolutional neural network (CNN) is the primary technique that has greatly promoted the development of computer vision technologies. However, there is little research on how to allocate parameters in different convolution layers when designing CNNs. We research mainly on revealing the relationship between CNN parameter distribution, i.e., the allocation of parameters in convolution layers, and the discriminative performance of CNN. Unlike previous works, we do not append more elements into the network, such as more convolution layers or denser short connections. We focus on enhancing the discriminative performance of CNN through varying its parameter distribution under strict size constraint. We propose an energy function to represent the CNN parameter distribution, which establishes the connection between the allocation of parameters and the discriminative performance of CNN. Extensive experiments with shallow CNNs on three public image classification data sets demonstrate that the CNN parameter distribution with a higher energy value will promote the model to obtain better performance. According to the motivated observation, the problem of finding the optimal parameter distribution can be transformed into an optimization problem of finding the biggest energy value. We present a simple yet effective guideline that uses balanced parameter distribution to design CNNs. Extensive experiments on ImageNet with three popular backbones, i.e., AlexNet, ResNet34, and ResNet101, demonstrate that the proposed guideline can make consistent improvements upon different baselines under strict size constraint.

We ranked third in the Predictive Analytics Competition (PAC) 2019 challenge by achieving a mean absolute error (MAE) of 3.33 years in predicting age from T1-weighted MRI brain images. Our approach combined seven algorithms that allow generating predictions when the number of features exceeds the number of observations, in particular, two versions of best linear unbiased predictor (BLUP), support vector machine (SVM), two shallow convolutional neural networks (CNNs), and the famous ResNet and Inception V1. Ensemble learning was derived from estimating weights via linear regression in a hold-out subset of the training sample. We further evaluated and identified factors that could influence prediction accuracy: choice of algorithm, ensemble learning, and features used as input/MRI image processing. Our prediction error was correlated with age, and absolute error was greater for older participants, suggesting to increase the training sample for this subgroup. Our results may be used to guide researchers to build age predictors on healthy individuals, which can be used in research and in the clinics as non-specific predictors of disease status.

Studies are being actively conducted on camera-based driver gaze tracking in a vehicle environment for vehicle interfaces and analyzing forward attention for judging driver inattention. In existing studies on the single-camera-based method, there are frequent situations in which the eye information necessary for gaze tracking cannot be observed well in the camera input image owing to the turning of the driver’s head during driving. To solve this problem, existing studies have used multiple-camera-based methods to obtain images to track the driver’s gaze. However, this method has the drawback of an excessive computation process and processing time, as it involves detecting the eyes and extracting the features of all images obtained from multiple cameras. This makes it difficult to implement it in an actual vehicle environment. To solve these limitations of existing studies, this study proposes a method that uses a shallow convolutional neural network (CNN) for the images of the driver’s face acquired from two cameras to adaptively select camera images more suitable for detecting eye position; faster R-CNN is applied to the selected driver images, and after the driver’s eyes are detected, the eye positions of the camera image of the other side are mapped through a geometric transformation matrix. Experiments were conducted using the self-built Dongguk Dual Camera-based Driver Database (DDCD-DB1) including the images of 26 participants acquired from inside a vehicle and the Columbia Gaze Data Set (CAVE-DB) open database. The results confirmed that the performance of the proposed method is superior to those of the existing methods.

BackgroundComputed tomography (CT) is widely used in clinical practice due to its ability to provide detailed anatomical information. However, variations in radiation dose can affect image quality, potentially compromising the performance and reliability of artificial intelligence (AI) models applied to these images.PurposeTo evaluate the robustness of radiomics‐based and deep learning‐based models to variations in CT dose levels using a standardized dataset obtained from a 3D‐printed anthropomorphic phantom simulating liver tissue with anomalies, as well as in the publicly available dataset CT‐ORG with real patient data for organ classification. This study is in an early experimental stage, tested only on retrospective data.MethodsA total of 1378 image series from 649 scans were acquired across 13 scanners from four manufacturers at five dose levels. Features were extracted from six regions of interest (ROIs), representing four liver tissue types (normal, cyst, hemangioma, metastasis), using four methods: PyRadiomics, a shallow convolutional neural network (CNN), SwinUNETR, and a CT foundation model (CT‐FM). Feature stability was assessed using the Intraclass Correlation Coefficient (ICC), while Uniform Manifold Approximation and Projection (UMAP) was employed to evaluate tissue types separability and the influence of scanner variations. Generalizability was tested by training liver tissue classifiers on one dose level and testing on others, alongside a dose classification task (10‐fold cross‐validation) to determine the sensitivity of each method to dose variations. In addition, we compared the four methods in addressing the task of organ classification (10‐fold cross‐validation) with the CT‐ORG dataset containing 140 CT scans acquired with varying dose levels.ResultsRadiomic features showed limited robustness to dose variations, leading to reduced performance in liver tissue classification and the lowest ICC among methods (ICC: 0.8355 ± 0.1705). SwinUNETR and CT‐FM exhibited the highest stability (SwinUNETR ICC: 0.9528 ± 0.0272; CT‐FM ICC: 0.9347 ± 0.0420), clearly above the Shallow CNN (ICC: 0.8416 ± 0.2018). CT‐FM also showed strong generalization across dose levels: its features effectively distinguished between liver tissue types and dose levels simultaneously, without compromising performance in either task. Consistent with these trends in dose sensitivity, CT‐FM obtained the highest dose‐classification accuracy (0.6517 ± 0.0179), whereas SwinUNETR showed the lowest (0.3796 ± 0.0250). These trends were confirmed in the context of organ classification with real patient data on the CT‐ORG dataset, where CT‐FM achieved the highest accuracy (0.965).ConclusionsThe study highlights the limited robustness of traditional radiomics and deep models to CT dose variation and underscores the potential of foundation models like CT‐FM to enable robust clinical applications by mitigating dose‐related variability. This enhanced performance is likely due to the model's pretraining on large and diverse datasets, allowing it to learn robust and generalizable representations across varying acquisition conditions.

This paper presents the BigEarthNet that is a new large-scale multi-label Sentinel-2 benchmark archive. The BigEarthNet consists of 590,326 Sentinel-2 image patches, each of which is a section of i) 120x120 pixels for 10m bands; ii) 60x60 pixels for 20m bands; and iii) 20x20 pixels for 60m bands. Unlike most of the existing archives, each image patch is annotated by multiple land-cover classes (i.e., multi-labels) that are provided from the CORINE Land Cover database of the year 2018 (CLC 2018). The BigEarthNet is significantly larger than the existing archives in remote sensing (RS) and thus is much more convenient to be used as a training source in the context of deep learning. This paper first addresses the limitations of the existing archives and then describes the properties of the BigEarthNet. Experimental results obtained in the framework of RS image scene classification problems show that a shallow Convolutional Neural Network (CNN) architecture trained on the BigEarthNet provides much higher accuracy compared to a state-of-the-art CNN model pre-trained on the ImageNet (which is a very popular large-scale benchmark archive in computer vision). The BigEarthNet opens up promising directions to advance operational RS applications and research in massive Sentinel-2 image archives.