Networks For Image Classification Research Articles

Analysis of Learnability of a Novel Hybrid Quantum-Classical Convolutional Neural Network in Image Classification

Abstract We design a new hybrid quantum-classical convolutional neural network (HQCCNN) model based on parameter quantum circuits. In this model, we use parameterized quantum circuits (PQC) to redesign the convolutional layer in classical convolutional neural networks (CNN), forming a new quantum convolutional layer to achieve unitary transformation of quantum states, enabling the model to more accurately extract hidden information from images. At the same time, we combine the classical fully connected layer with PQC to form a new hybrid quantum-classical fully connected layer to further improve the accuracy of classification. Finally, we used the MNIST dataset to test the potential of HQCCNN. The results indicate that HQCCNN has good performance in solving classification problems. In binary classification tasks, the classification accuracy of numbers 5 and 7 is as high as 99.71%. And in multivariate classification, the accuracy rate also reached 98.51%. Finally, we compare the performance of HQCCNN with other models and find that HQCCNN has better classification performance and convergence speed.

Multiscale dilated dense network for hyperspectral image classification

Hyperspectral image has rich spectral information and can accurately reflect the real condition of ground objects, so it is widely used. However, there are huge challenges in hyperspectral image, such as few training samples, the phenomena of “the foreign matter same spectrum” and “the same thingdifferent spectrum” can not be distinguished. The hyperspectral image classification method based on convolutional neural network can solve the above problems and achieve good classification performance. Most classification methods based on convolutional neural network only extract the deep semantic information at the end of the network, ignoring the shallow details. At the same time, Methods of this type increase the receptive field by designing deep network structure, which is prone to gradient vanishing. To solve the above two problems, a multiscale dilated dense network is proposed for hyperspectral image classification. Firstly, hyperspectral data cubes with different neighborhood sizes are selected as network inputs to construct a multi-channel network structure with different scale inputs. Secondly, a multiscale feature mapping module with fusion of dilated convolution is designed. Different numbers of dilated convolution are used to increase the model's receptive field without increasing network parameters, and spatial-spectral features of different scales are fused. Skip connections are added between the designed modules to fully combine the shallow and deep features of the network, without skip connections in the modules, reduce the computation of the overall network. Finally, the resulting features are input into the fully connected layer and softmax to complete the classification. In addition, in order to prevent the network from overfitting, we add dropout regularization method after the fully connected layer to improve the generalization ability of the model. Experiments are conducted on three publicly available datasets, Indian Pines, University of Pavia and Salinas, with accuracy rates of 99.28%, 99.48% and 99.25%, respectively. The experimental results show that compared with the classification methods based on convolutional neural network, the proposed method can effectively extract multiscale features of hyperspectral image and enhance the expression ability of spatial-spectral fusion features, and the classification performance is better.

Convolutional Neural Networks for Medical Image Segmentation and Classification: A Review

Convolutional Neural Networks for Medical Image Segmentation and Classification: A Review

Recognition of edible mushrooms based on different image classification methods

The mushroom is one of the most common foods in nature, but there are many different types and many of them are deadly. The aim of this paper is to use computer vision methods to be able to provide a range of assistance to people on safari in identifying mushrooms and the future development of mushroom poisoning and non-poisoning identification. Considering the superior performance of convolutional neural networks in image recognition and classification, the advantages of multi-layer perceptrons and deep structure make it possible to achieve high recognition accuracy, three different CNN network models, GoogLeNet ResNet and SENet were chosen. They have been used to differentiate between toxic and non-poisonous mushrooms on 11,000 mushrooms. The tests showed that SENet had the greatest results, with a recognition accuracy of over 90%. With a recognition accuracy of around 65% each, GoogLeNet and ResNet placed second and third, respectively.

Quantized Graph Neural Networks for Image Classification

Researchers have resorted to model quantization to compress and accelerate graph neural networks (GNNs). Nevertheless, several challenges remain: (1) quantization functions overlook outliers in the distribution, leading to increased quantization errors; (2) the reliance on full-precision teacher models results in higher computational and memory overhead. To address these issues, this study introduces a novel framework called quantized graph neural networks for image classification (QGNN-IC), which incorporates a novel quantization function, Pauta quantization (PQ), and two innovative self-distillation methods, attention quantization distillation (AQD) and stochastic quantization distillation (SQD). Specifically, PQ utilizes the statistical characteristics of distribution to effectively eliminate outliers, thereby promoting fine-grained quantization and reducing quantization errors. AQD enhances the semantic information extraction capability by learning from beneficial channels via attention. SQD enhances the quantization robustness through stochastic quantization. AQD and SQD significantly improve the performance of the quantized model with minimal overhead. Extensive experiments show that QGNN-IC not only surpasses existing state-of-the-art quantization methods but also demonstrates robust generalizability.

A novel approach for tomato leaf disease classification with deep convolutional neural networks

Computer-aided automation systems that detect plant diseases are one of the challenging research areas that provide effective results in the agricultural field. Tomato crops are a major product with high commercial value worldwide and are produced in large quantities. This study proposes a new approach for the automatic detection of tomato leaf diseases, which employs classical learning methods and deep neural networks for image classification. Specifically, Local Binary Pattern (LBP) method was used for feature extraction in classical learning methods, while Extreme Learning Machines, k-Nearest Neighborhood (kNN), and Support Vector Machines (SVM) were used for classification. On the other hand, a novel Convolutional Neural Network (CNN) framework with its parameters and layers was employed for deep learning. The study shows that the accuracy values obtained from the proposed approach are better than the state-of-the-art studies. The classification process was carried out with different numbers of classes, including binary classification (healthy vs. unhealthy), 6-class, and 10-class classification for distinguishing different types of diseases. The results indicate that the CNN model outperforms classical learning methods, with accuracy values of 99.5%, 98.50%, and 97.0% obtained for the classification of 2, 6, and 10 classes, respectively. In future studies, computer-aided automated systems can be utilized to detect different diseases for various plant species.

Adaptive fuzzy convolutional neural network for medical image classification

Due to resource constraints and the diverse nature of the devices involved, energy efficiency and scalability enhancement are important challenges in the Internet of Things (IoT) ecosystem. It is difficult to manage the edge resources in a consistent way that promotes cooperation and sharing of resources across the devices because of the heterogeneity of the Internet of Things devices and the dynamic nature of the surroundings in which edge computing takes place. In this research, we offer Intelligent techniques for resource optimization for Internet of Things devices. This is a full-stack system architecture to support across heterogeneous Internet of Things devices that have limited resources. The paradigm that is being suggested is made up of several edge servers, and Internet of Things (IoT) devices have the qualities of being heterogeneity-compatible, high performing, and intelligently adaptable. In order to do this, a clustered environment is generated in heterogeneous Internet of Things devices, and a routing method called Search and Rescue Optimization is used to set up connections between the CH nodes. After that, the edge nodes that are closest to the source of the data are chosen for transmission. Overall, what was suggested Multi-Edge-IoT accomplishes superior efficiency in terms of consumption of energy, latency, communication overhead, and packet loss rate than existing approaches to attaining energy efficiency in the Internet of Things.

Spectral unmixing-based Arctic plant species analysis using a spectral library and terrestrial hyperspectral Imagery: A case study in Adventdalen, Svalbard

Remote sensing is an invaluable tool for monitoring the rapid changes in Arctic vegetation distribution caused by global warming. Although hyperspectral data consisting of contiguous spectral bands enables the quantitative analysis of remote sensing data, mapping Arctic vegetation using hyperspectral remote sensing remains challenging due to the difficulty in acquiring data from the region. Additionally, the mixed pixel issue, which represents a mixture of more than one plant species due to low-spatial resolution, hinders accurate mapping of Arctic vegetation. To address these limitations, we collected hyperspectral information on the dominant plant species, such as shrubs and graminoids, in Adventdalen, Svalbard, and investigated effective methods for mapping Arctic vegetation using spectral unmixing. First, labeled datasets were constructed for Arctic plant species by extracting pixel data from terrestrial hyperspectral images. A spectral library was developed using the labeled datasets and used for spectral unmixing as endmembers. We employed three established classifiers, random forest, support vector machine, and one-dimensional convolutional neural network (1D-CNN) for hyperspectral image classification. Subsequently, we quantitatively and qualitatively compared the classification performances of these machine learning-based classifiers to determine the optimal classification method for validation purposes. The first derivative spectra of smoothed reflectance (RS,FD) and the 1D-CNN classifier were used to identify ground truth by classifying pixels of Arctic vegetation because they achieved the highest statistical accuracies of 0.9892 (Kappa = 0.9880) and 0.9352 (Kappa = 0.9280) for the two independent test sets and produced the most accurate vegetation maps. The spectral library using the RS,FD spectrum showed high spectral discriminability and potential for estimating the abundance of classes from simulated mixed pixels, resulting in good agreement with the ground truth. Accordingly, our findings provide valuable insights and references for large-scale mapping studies using remote sensing data and offer effective monitoring strategies for Arctic vegetation changes in response to climate change.

Multiple vision architectures-based hybrid network for hyperspectral image classification

Multiple vision architectures-based hybrid network for hyperspectral image classification

Anomaly detection and visualization for sound data with pre-trained deep neural networks

Anomaly detection in the industrial domain is a critical issue. Although anomaly detection needs expert's experiences or huge person-hours, recent researches of deep neural networks can save much labor. This paper proposes an anomaly detection method for sound data with pre-trained deep neural networks, and an anomaly visualization method to specify the frequency components of anomalous sounds. Common anomaly detection methods train deep neural networks which can detect anomalies in specific domains. However, training such networks from scratch costs a lot because the neural networks have numerous parameters and use lots of data to improve the performance. To address this problem, anomaly detection using the pre-trained image classification networks has been proposed. This method uses feature representations of classification networks and the Mahalanobis distance as the anomaly scores. In this paper, we propose a method which can apply the above method to sound data and evaluate its performance with a dataset for industrial sound which contains machine sounds under normal and anomalous operating conditions in real factory environments. We also propose a visualization method to specify the frequency components when abnormal sounds occur.

Dynamic Hand Gesture Recognition for Indian Sign Language using Integrated CNN-LSTM Architecture

Human Centered Computing is an emerging research field that aims to understand human behavior. Dynamic hand gesture recognition is one of the most recent, challenging and appealing application in this field. We have proposed one vision based system to recognize dynamic hand gestures for Indian Sign Language (ISL) in this paper. The system is built by using a unified architecture formed by combining Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM). In order to hit the shortage of a huge labeled hand gesture dataset, we have created two different CNN by retraining a well known image classification networks GoogLeNet and VGG16 using transfer learning. Frames of gesture videos are transformed into features vectors using these CNNs. As these videos are prearranged series of image frames, LSTM model have been used to join with the fully-connected layer of CNN. We have evaluated the system on three different datasets consisting of color videos with 11, 64 and 8 classes. During experiments it is found that the proposed CNN-LSTM architecture using GoogLeNet is fast and efficient having capability to achieve very high recognition rates of 93.18%, 97.50%, and 96.65% on the three datasets respectively.

Re-abstraction and perturbing support pair network for few-shot fine-grained image classification

The goal of few-shot fine-grained image classification (FSFGIC) is to distinguish subordinate-level categories with subtle visual differences such as the species of bird and models of car with only a few samples. In this work, we argue that a designed network that has the ability to better distinguish feature descriptors of different categories will effectively improve the performance of FSFGIC. We propose a re-abstraction and perturbing support pair network (RaPSPNet) for FSFGIC. Specifically, we first design a feature re-abstraction embedding (FRaE) module which can not only effectively amplify the difference between the feature information from different categories but also better extract the feature information from images. Furthermore, a novel perturbing support pair (PSP) based similarity measure module is designed which evaluates the relationships of feature information among a query image and two different categories of support images (a support pair) at the same time for guiding the designed FRaE module to find salient feature information from the same category of query and support images and find distinguishable feature information from the different categories of query and support images. Extensive experiments on FSFGIC tasks demonstrate the superiority of the proposed methods over state-of-the-art benchmarks.

A comparative analysis of various activation functions and optimizers in a convolutional neural network for hyperspectral image classification

A comparative analysis of various activation functions and optimizers in a convolutional neural network for hyperspectral image classification

Adaptive Multi-Feature Fusion Graph Convolutional Network for Hyperspectral Image Classification

Graph convolutional networks (GCNs) are a promising approach for addressing the necessity for long-range information in hyperspectral image (HSI) classification. Researchers have attempted to develop classification methods that combine strong generalizations with effective classification. However, the current HSI classification methods based on GCN present two main challenges. First, they overlook the multi-view features inherent in HSIs, whereas multi-view information interacts with each other to facilitate classification tasks. Second, many algorithms perform a rudimentary fusion of extracted features, which can result in information redundancy and conflicts. To address these challenges and exploit the strengths of multiple features, this paper introduces an adaptive multi-feature fusion GCN (AMF-GCN) for HSI classification. Initially, the AMF-GCN algorithm extracts spectral and textural features from the HSIs and combines them to create fusion features. Subsequently, these three features are employed to construct separate images, which are then processed individually using multi-branch GCNs. The AMG-GCN aggregates node information and utilizes an attention-based feature fusion method to selectively incorporate valuable features. We evaluated the model on three widely used HSI datasets, i.e., Pavia University, Salinas, and Houston-2013, and achieved accuracies of 97.45%, 98.03%, and 93.02%, respectively. Extensive experimental results show that the classification performance of the AMF-GCN on benchmark HSI datasets is comparable to those of state-of-the-art methods.

H-CapsNet: A capsule network for hierarchical image classification

In this paper, we present H-CapsNet, a capsule network for hierarchical image classification. Our network makes use of the natural capacity of CapsNets (capsule networks) to capture hierarchical relationships. Thus, our network is such that each multi-layer capsule network accounts for each of the class hierarchies using dedicated capsules. Further, we make use of a modified hinge loss that enforces consistency amongst the hierarchies involved. We also present a strategy to dynamically adjust the training parameters to achieve a better balance between the class hierarchies under consideration. We have performed experiments using several widely available datasets and compared them against several alternatives. In our experiments, H-CapsNet delivers a margin of improvement over competing hierarchical classification networks elsewhere in the literature.

Application of Swin-Unet for pointer detection and automatic calculation of readings in pointer-type meters

Pointer-type meters are widely used in military, industrial, and aerospace applications. In this paper, we propose a method to automatically calculate the readings of pointer-type meters that exhibits strong robustness to various situations, such as complex backgrounds, tilted meters, blurred images, and meter images with uneven illumination. First, the mask maps of scale and pointers are obtained using the Swin-Unet semantic segmentation network. For the mask map of scale, the Swin Transformer image classification network is used to identify the values of the scale and the coordinates of the centroids of the scale, and for the mask map of the pointer, the least skeleton circle method is proposed to fit the linear equation of the pointer. Second, the influence of the pointer and several adjacent scale on the meter reading is considered, and the weighted angle method (WAM) is proposed to calculate the meter reading. In order to verify the robustness of the algorithm in this paper, the pointer detection method is compared with the traditional pointer detection method, and it is found that the pointer detection method in this paper works better, and the pointer detection algorithm in this paper is applied to different semantic segmentation results, and it is verified that the pointer detection algorithm in this paper can be well adapted to different semantic segmentation results. This paper also compares the proposed algorithm with the existing meter reading calculation methods. The experiments show that using WAM on uncorrected meter images reduces the error by 30% compared with the traditional angle method, and using WAM on corrected meter images reduces the error by about 50%, which finally verifies the effectiveness of the algorithm in this paper.

Introducing shape priors in Siamese networks for image classification

The efficiency of deep neural networks is increasing, and so is the amount of annotated data required for training them. We propose a solution improving the learning process of a classification network with less labeled data. Our approach is to inform the classifier of the elements it should focus on to make its decision by supplying it with some shape priors. These shape priors are expressed as binary masks, giving a rough idea of the shape of the relevant elements for a given class. We resort to Siamese architecture and feed it with image/mask pairs. By inserting shape priors, only the relevant features are retained. This provides the network with significant generalization power without requiring a specific domain adaptation step. This solution is tested on some standard cross-domain digit classification tasks and on a real-world video surveillance application. Extensive tests show that our approach outperforms the classical classifier by generating a good latent space with less training data. Code is available at https://github.com/halqasir/MG-Siamese.

A Comprehensive Survey on Model Quantization for Deep Neural Networks in Image Classification

Recent advancements in machine learning achieved by Deep Neural Networks (DNNs) have been significant. While demonstrating high accuracy, DNNs are associated with a huge number of parameters and computations, which leads to high memory usage and energy consumption. As a result, deploying DNNs on devices with constrained hardware resources poses significant challenges. To overcome this, various compression techniques have been widely employed to optimize DNN accelerators. A promising approach is quantization, in which the full-precision values are stored in low bit-width precision. Quantization not only reduces memory requirements but also replaces high-cost operations with low-cost ones. DNN quantization offers flexibility and efficiency in hardware design, making it a widely adopted technique in various methods. Since quantization has been extensively utilized in previous works, there is a need for an integrated report that provides an understanding, analysis, and comparison of different quantization approaches. Consequently, we present a comprehensive survey of quantization concepts and methods, with a focus on image classification. We describe clustering-based quantization methods and explore the use of a scale factor parameter for approximating full-precision values. Moreover, we thoroughly review the training of a quantized DNN, including the use of a straight-through estimator and quantization regularization. We explain the replacement of floating-point operations with low-cost bitwise operations in a quantized DNN and the sensitivity of different layers in quantization. Furthermore, we highlight the evaluation metrics for quantization methods and important benchmarks in the image classification task. We also present the accuracy of the state-of-the-art methods on CIFAR-10 and ImageNet. This article attempts to make the readers familiar with the basic and advanced concepts of quantization, introduce important works in DNN quantization, and highlight challenges for future research in this field.

Recognition of Additive Manufacturing Parts Based on Neural Networks and Synthetic Training Data: A Generalized End-to-End Workflow

Additive manufacturing (AM) is becoming increasingly relevant among established manufacturing processes. AM parts must often be recognized to sort them for part- or order-specific post-processing. Typically, the part recognition is performed manually, which represents a bottleneck in the AM process chain. To address this challenge, a generalized end-to-end workflow for automated visual real-time recognition of AM parts is presented, optimized, and evaluated. In the workflow, synthetic training images are generated from digital AM part models via rendering. These images are used to train a neural network for image classification, which can recognize the printed AM parts without design adaptations. As each production batch can consist of new parts, the workflow is generalized to be applicable to individual batches without adaptation. Data generation, network training and image classification are optimized in terms of the hardware requirements and computational resources for industrial applicability at low cost. For this, the influences of the neural network structure, the integration of a physics simulation in the rendering process and the total number of training images per AM part are analyzed. The proposed workflow is evaluated in an industrial case study involving 215 distinct AM part geometries. Part classification accuracies of 99.04% (top three) and 90.37% (top one) are achieved.

Deep concatenated features with improved heuristic-based recurrent neural network for hyperspectral image classification

Deep concatenated features with improved heuristic-based recurrent neural network for hyperspectral image classification