Visual Image Classification Research Articles

Visual Ship Image Synthesis and Classification Framework Based on Attention-DCGAN

To improving ship image generation and classification tasks, a deep convolution generative adversarial network based on attention mechanism (ADCGAN) model was constructed. The rectified linear unit (ReLU) activation function was adopted, and three Deconv layers and Conv layers were added to both the generator and discriminator. Subsequently, an attention mechanism was added to the generator, while spectral normalization (SN) was added to the discriminator. Mean squared error (MSE) was used as loss function to stabilize the training process. Furthermore, ship classification tasks were performed using the generated ship images by end-to-end training of the classification network, enabling ship data augmentation and co-learning with other tasks. Experimental results on the Ship700 and Seaship7000 datasets demonstrate that the ADCGAN model can generate clear and robust ship images, with PSNR, LIPIPS, MS-SSIM values of 20.279 and 27.523, 0.596 and 0.096, 0.781 and 0.947, respectively. The effectiveness of the proposed method in ship image classification tasks was also verified, providing a data foundation for other collaborative tasks.

A RESEARCH OF THE LATEST APPROACHES TO VISUAL IMAGE RECOGNITION AND CLASSIFICATION

Context. The paper provides an overview of current methods for recognizing and classifying visual images in static images or video stream. The paper will discuss various approaches, including machine learning, current problems of these methods and possible improvements. The biggest challenges of the visual image retrieval and classification task are discussed. The main emphasis is placed on the review of such promising algorithms as SSD, YOLO, R-CNN, an overview of the principles of these methods, network architectures.
 Objective. The aim of the work is to analyze existing studies and find the best algorithm for recognizing and classifying visual images for further activities.
 Method. Primary method is to compare different factors of algorithms in order to select the most perspective one. There are different marks to compare, like image processing speed, accuracy. There are a number of studies and publications that propose methods and algorithms for solving the problem of finding and classifying images in an image [3–6]. It should be noted that most promising approaches are based on machine learning methods. It is worth noting that the proposed methods have drawbacks due to the imperfect implementation of the Faster R-CNN, YOLO, SSD algorithms for working with streaming video. The impact of these drawbacks can be significantly reduced by applying the following solutions: development of combined identification methods, processing of edge cases – tracking the position of identified objects, using the difference between video frames, additional preliminary preparation of input data. Another major area for improvement is the optimization of methods to work with real-time video data, as most current methods focus on images.
 Results. As an outcome of the current research we have found an optimal algorithm for further researches and optimizations.
 Conclusions. Analysis of existent papers and researches has demonstrated the most promising algorithm for further optimizations and experiments. Also current approaches still have some space for further. The next step is to take the chosen algorithm and investigate possibilities to enhance it.

Correction of Intensity of Bone Involvement (IBI) Index by Lean Body Mass and Body Surface Area

The Intensity of Bone Involvement (IBI) parameter is utilized to quantitatively assess bone involvement in PET/CT images of multiple myeloma (MM) patients, providing an objective and reproducible analytical approach. This study proposes the mathematical correction of IBI for lean body mass (LBM) and body surface area (BSA) in 18F-FDG PET/CT images of MM patients. Beyond the corrections, our aim is to evaluate the correlation of corrected IBIs and visual image analysis, along with their relationship with overall patient survival. Methods: This retrospective study analyzed 18F-FDG PET/CT images and clinical data from 73 MM patients. LBM was calculated using the James and Janma models, while BSA was calculated using the DuBois model. The conventionally calculated IBI parameter was then corrected for LBM and BSA. Three experienced nuclear medicine physicians visually classified the intensity of bone involvement as: negative, mild, moderate, marked, and very marked. We compared standard IBI and corrected indices with visual classification using Kruskal-Wallis test and Dunn's post-hoc test. Cox proportional hazards regression analysis was employed to test the association with overall patient survival. Statistical significance was established for p-values <0.05. Results: A positive relationship was observed between both standard IBI and its corrected values with visual image classification. No statistically significant association was found between IBI values and overall patient survival. Conclusions: IBI corrected for both LBM or BSA exhibited similar performance to standard IBI compared to visual image classification. However, IBI values (both standard and its corrections) were not associated with overall survival (OS) in our study.

The manipulative essence of visual media: from man-made images to generative deepfakes

The paper explores the problems of manipulation and reliability of information in the visual media. Manipulation is part of their essence. Distortions in the message arise from a shift in attention from reality as such to how it’s seen by others or who and how it’s represented. The objective vision is replaced by a presentation of the vision of others in visual media. This opens up the possibility of manipulation and becomes the reason for the creation of visual falsifications. Modernity has become the era of post-truth. Visual media are increasingly shifting towards total falsification with the development of digital technologies, artificial intelligence. The generated image turns out to be more real and reliable than reality itself. This paper uses the classification of visual media images based on how they are created. An attempt is made to identify the causes of distortions in the representation of reality at different stages of the development of visual media. The article analyzes how the technologies of creating visual images affect the development of manipulative and falsifying practices that undermine the credibility of the visual message.

Hierarchical Multi-Attention Transfer for Knowledge Distillation

Knowledge distillation (KD) is a powerful and widely applicable technique for the compression of deep learning models. The main idea of knowledge distillation is to transfer knowledge from a large teacher model to a small student model, where the attention mechanism has been intensively explored in regard to its great flexibility for managing different teacher-student architectures. However, existing attention-based methods usually transfer similar attention knowledge from the intermediate layers of deep neural networks, leaving the hierarchical structure of deep representation learning poorly investigated for knowledge distillation. In this paper, we propose a hierarchical multi-attention transfer framework (HMAT) , where different types of attention are utilized to transfer the knowledge at different levels of deep representation learning for knowledge distillation. Specifically, position-based and channel-based attention knowledge characterize the knowledge from low-level and high-level feature representations, respectively, and activation-based attention knowledge characterize the knowledge from both mid-level and high-level feature representations. Extensive experiments on three popular visual recognition tasks, image classification, image retrieval, and object detection, demonstrate that the proposed hierarchical multi-attention transfer or HMAT significantly outperforms recent state-of-the-art KD methods.

A hybrid deep learning framework for automated visual image classification using EEG signals

A hybrid deep learning framework for automated visual image classification using EEG signals

Automated Classification of Brain Tumors from Magnetic Resonance Imaging Using Deep Learning.

Brain tumor classification is crucial for medical evaluation in computer-assisted diagnostics (CAD). However, manual diagnosis of brain tumors from magnetic resonance imaging (MRI) can be time-consuming and complex, leading to inaccurate detection and classification. This is mainly because brain tumor identification is a complex procedure that relies on different modules. The advancements in Deep Learning (DL) have assisted in the automated process of medical images and diagnostics for various medical conditions, which benefits the health sector. Convolutional Neural Network (CNN) is one of the most prominent DL methods for visual learning and image classification tasks. This study presents a novel CNN algorithm to classify the brain tumor types of glioma, meningioma, and pituitary. The algorithm was tested on benchmarked data and compared with the existing pre-trained VGG16, VGG19, ResNet50, MobileNetV2, and InceptionV3 algorithms reported in the literature. The experimental results have indicated a high classification accuracy of 98.04%, precision, recall, and f1-score success rate of 98%, respectively. The classification results proved that the most common kinds of brain tumors could be categorized with a high level of accuracy. The presented algorithm has good generalization capability and execution speed that can be helpful in the field of medicine to assist doctors in making prompt and accurate decisions associated with brain tumor diagnosis.

Research on spiking neural network in art visual image classification

Research on spiking neural network in art visual image classification

Research on spiking neural network in art visual image classification

Research on spiking neural network in art visual image classification

Artificial Intelligence to Aid Glaucoma Diagnosis and Monitoring: State of the Art and New Directions.

Recent developments in the use of artificial intelligence in the diagnosis and monitoring of glaucoma are discussed. To set the context and fix terminology, a brief historic overview of artificial intelligence is provided, along with some fundamentals of statistical modeling. Next, recent applications of artificial intelligence techniques in glaucoma diagnosis and the monitoring of glaucoma progression are reviewed, including the classification of visual field images and the detection of glaucomatous change in retinal nerve fiber layer thickness. Current challenges in the direct application of artificial intelligence to further our understating of this disease are also outlined. The article also discusses how the combined use of mathematical modeling and artificial intelligence may help to address these challenges, along with stronger communication between data scientists and clinicians.

Feature Extraction Method of Art Visual Communication Image Based on 5G Intelligent Sensor Network

5G intelligent sensor network technology realizes the perception, processing, and transmission of information. It forms the three pillars of information technology together with computer technology and communication technology and is an important part of the Internet of Things technology. The 5G smart sensor network is a wireless communication module added to the sensor nodes, and a wireless communication network is formed by a large number of stationary or movable sensor nodes in the form of self-organization and multihop transmission. This paper proposes a keypoint feature extraction method based on deep learning, which can extract keypoint local features for matching. This method uses the convolutional network structure, which is pretrained based on the Siamese network structure and then adjusted to the ternary network structure to continue training to improve the accuracy. This paper proposes a high-art visual communication image classification based on multifeature extraction and classification decision fusion. In the data preprocessing stage, the correlation alignment algorithm is performed on the datasets of different domains (source domain and target domain) to reduce the difference in spatial distribution, and then, a multifeature extractor is designed to extract artistic visual communication images and spatial information. In the process, the multitask learning method is introduced to jointly train the networks of multiple data sets to reduce the degree of overfitting of the model, solve the problem of insufficient labeled samples in the target domain data set, and affect the classification accuracy of high-art visual communication images. Finally, the classification results are obtained through the fusion of voting decisions. The experimental results show that the advantage of this framework is that it utilizes the artistic visual communication image and spatial structure information from the source and target scenes, which can significantly reduce the dependence on the number of labeled samples in the target domain and improve the classification performance. In this paper, a dual-channel deep residual convolutional neural network is designed. The multiple convolution layers of the residual module in the network use hard parameters to share, so that the deep feature representation on the joint spatial spectrum dimension can be automatically extracted. The features extracted by the network are transferred to maximize the auxiliary role of the labeled samples in the source domain and avoid the negative transfer problem caused by the forced transfer between irrelevant samples.

Differential Screening of Herniated Lumbar Discs Based on Bag of Visual Words Image Classification Using Digital Infrared Thermographic Images.

Doctors in primary hospitals can obtain the impression of lumbosacral radiculopathy with a physical exam and need to acquire medical images, such as an expensive MRI, for diagnosis. Then, doctors will perform a foraminal root block to the target root for pain control. However, there was insufficient screening medical image examination for precise L5 and S1 lumbosacral radiculopathy, which is most prevalent in the clinical field. Therefore, to perform differential screening of L5 and S1 lumbosacral radiculopathy, the authors applied digital infrared thermographic images (DITI) to the machine learning (ML) algorithm, which is the bag of visual words method. DITI dataset included data from the healthy population and radiculopathy patients with herniated lumbar discs (HLDs) L4/5 and L5/S1. A total of 842 patients were enrolled and the dataset was split into a 7:3 ratio as the training algorithm and test dataset to evaluate model performance. The average accuracy was 0.72 and 0.67, the average precision was 0.71 and 0.77, the average recall was 0.69 and 0.74, and the F1 score was 0.70 and 0.75 for the training and test datasets. Application of the bag of visual words algorithm to DITI classification will aid in the differential screening of lumbosacral radiculopathy and increase the therapeutic effect of primary pain interventions with economical cost.

Cross-Media Intelligent Perception and Retrieval Analysis Application Technology Based on Deep Learning Education

This paper proposes a research on cross-media intelligent perception and retrieval analysis technology based on deep learning education in view of the complex learning knowledge environment of artificial intelligence of traditional cross-media retrieval technology, which is unable to obtain retrieval information timely and accurately. Based on the cross-media theory, this paper analyzes the pre-processing of cross-media data, transforms from a single media form such as text, voice, image, and video to a cross-media integration covering network space and physical space, and designs a cross-media intelligent perception platform system. With multi-core method of typical correlation analysis algorithm, this paper develops a new joint learning framework based on the joint feature selection and subspace learning cross-modal retrieval. What’s more, the algorithm is tested experimentally. The results show that the retrieval analysis technology is significantly better than the traditional media retrieval technology, which can effectively identify text semantics and visual image classification, and can better maintains the relevance of data content and the consistency of semantic information, which has important reference value for cross-media applications.

A Novel Luminance-Based Algorithm for Classification of Semi-Dark Images

Image classification of a visual scene based on visibility is significant due to the rise in readily available automated solutions. Currently, there are only two known spectrums of image visibility i.e., dark, and bright. However, normal environments include semi-dark scenarios. Hence, visual extremes that will lead to the accurate extraction of image features should be duly discarded. Fundamentally speaking there are two broad methods to perform visual scene-based image classification, i.e., machine learning (ML) methods and computer vision methods. In ML, the issues of insufficient data, sophisticated hardware and inadequate image classifier training time remain significant problems to be handled. These techniques fail to classify the visual scene-based images with high accuracy. The other alternative is computer vision (CV) methods, which also have major issues. CV methods do provide some basic procedures which may assist in such classification but, to the best of our knowledge, no CV algorithm exists to perform such classification, i.e., these do not account for semi-dark images in the first place. Moreover, these methods do not provide a well-defined protocol to calculate images’ content visibility and thereby classify images. One of the key algorithms for calculation of images’ content visibility is backed by the HSL (hue, saturation, lightness) color model. The HSL color model allows the visibility calculation of a scene by calculating the lightness/luminance of a single pixel. Recognizing the high potential of the HSL color model, we propose a novel framework relying on the simple approach of the statistical manipulation of an entire image’s pixel intensities, represented by HSL color model. The proposed algorithm, namely, Relative Perceived Luminance Classification (RPLC) uses the HSL (hue, saturation, lightness) color model to correctly identify the luminosity values of the entire image. Our findings prove that the proposed method yields high classification accuracy (over 78%) with a small error rate. We show that the computational complexity of RPLC is much less than that of the state-of-the-art ML algorithms.

Landslide inventory and susceptibility assessment using multiple statistical approaches along the Karakoram highway, northern Pakistan

China-Pakistan Economic Corridor (CPEC) is a framework of regional connectivity, which will not only benefit China and Pakistan but will have positive impact on Iran, Afghanistan, India, Central Asian Republic, and the region. The surrounding area in CPEC is prone to frequent disruption by geological hazards mainly landslides in northern Pakistan. Comprehensive landslide inventory and susceptibility assessment are rarely available to utilize for landslide mitigation strategies. This study aims to utilize the high-resolution satellite images to develop a comprehensive landslide inventory and subsequently develop landslide susceptibility maps using multiple techniques. The very high-resolution (VHR) satellite images are utilized to develop a landslide inventory using the visual image classification techniques, historic records and field observations. A total of 1632 landslides are mapped in the area. Four statistical models i.e., frequency ratio, artificial neural network, weights of evidence and logistic regression were used for landslide susceptibility modeling by comparing the landslide inventory with the topographic parameters, geological features, drainage and road network. The developed landslides susceptibility maps were verified using the area under curve (AUC) method. The prediction power of the model was assessed by the prediction rate curve. The success rate curves show 93%, 92.8%, 92.7% and 87.4% accuracy of susceptibility maps for frequency ratio, artificial neural network, weights of evidence and logistic regression, respectively. The developed landslide inventory and susceptibility maps can be used for land use planning and landslide mitigation strategies.

Exploiting structured high-level knowledge for domain-specific visual classification

In the last decade, deep learning models have yielded impressive performance on visual object recognition and image classification. However these methods still rely on learning visual data distributions and show difficulties in dealing with complex scenarios where visual appearance only is not enough to effectively tackle them. This is the case, for instance, of fine-grained image classification in domain-specific applications for which it is very complex to employ data-driven models because of the lack of large amounts of samples and that, instead, can be solved by resorting to specialized human knowledge. However, encoding this specialized knowledge and injecting it into deep models is not trivial.In this paper, we address this problem by: a) employing computational ontologies to model specialized knowledge in a structured representation and, b) building a hybrid visual-semantic classification framework. The classification method performs inference over a Bayesian Network graph, whose structure depends on the knowledge encoded in an ontology and evidences are built using the outputs of deep networks.We test our approach on a fine-grained classification task, employing an extremely complex dataset containing images from several fruit varieties as well as visual and semantic annotations. Since the classification is done at the variety level (e.g., discriminating between different cherry varieties), appearance changes slightly and expert domain knowledge — making using of contextual information — is required to perform classification accurately.Experimental results show that our approach significantly outperforms standard deep learning–based classification methods over the considered scenario as well as existing methods leveraging semantic information for classification. These results demonstrate, on one hand, the difficulty of purely-visual deep methods in tackling small and highly-specialized datasets and, on the other hard, the capabilities of our approach to effectively encode and use semantic knowledge for enhanced accuracy.

The use of sentinel 2A imageries to improve mangrove inventarization at coremap CTI monitoring areas

In 2014, the project preparation for Coral Reef Management (Coremap) CTI claim that mangrove at Selayar was approximately 676.7 hectares distributed at 15 villages. Tambolongan at Bontosikuyu district was the largest area of about 132 hectares. These values extracted using Landsat imageries based on visual and composite image classification. Since Sentinel 2A launched in 2015, the use of this satellite imagery has spread widely in ecosystem monitoring due to its higher spatial, spectral and temporal resolution and mostly because it is free to access. Nevertheless, field sampling is still needed to crosscheck the final result. Since 2015 three mangrove locations at Selayar island monitored annually. By 2018, two new locations at Pasi island added. In 2019 three sites added at two different islands, so now a total nine mangrove station covering the whole Selayar islands. Nine species of mangroves were found by the end of 2019 monitoring, which are Avicennia alba, A. officinalis, A. marina, Bruguiera gymnorrhiza, Rhizophora apiculata, R. mucronata, R. stylosa, Cerios tagal, and Sonneratia alba. Spectral image characterization by using 12 bands of Sentinel 2A shows that most of these mangroves had peak reflectances at 783 and 865 nanometer, and the highest reflectance are lower than 0.45. These findings may contribute to better spatial identifications for mangrove monitoring and also the consequences to the revision of previous information extracted from Landsat imageries.

Reading into the mind’s eye: Boosting automatic visual recognition with EEG signals

Classifying visual information is an apparently simple and effortless task in our everyday routine, but can we automatically predict what we see from signals emitted by the brain?While other researchers have already attempted to answer this question, we are the first to show that a commercially available BCI could be effectively used for visual image classification in real-world scenarios – when testing takes place at a completely different time than training data collection. The task is difficult, as it requires relating the noisy and low-level EEG signals to complex and highly semantic visual categories. In this paper, we propose different learning approaches and show that simpler classifiers such as Ridge Regression with Gabor filtering of the input EEG signal could be more effective than the powerful Long Short Term Memory Networks and Convolutional Neural Networks in this case of limited and noisy training data. We analyzed the importance of each electrode for the visual classification task and noticed that the sensors with the highest accuracy were the ones that recorded brain activity from regions known to be correlated more with higher level recognition and cognitive processes and less to lower-level visual signal processing. The result is also in accordance with research in computer vision with deep neural networks, which shows that semantic visual features are learned only at higher levels of neural depth.While EEG signals are weaker by themselves for the task of visual classification, we demonstrate that they could be powerful when combined with deep visual features extracted from the image, improving performance from 91% to over 97% in a multi-class recognition setting. Our tests show that EEG input brings additional information that is not learned by artificial deep networks on the given image training set. Thus, a commercially available BCI could be effectively used in conjunction with a deep learning based vision system to form together a stronger visual recognition system that is suitable for real-world applications.

Research on cooperative classification of multimedia visual images based on deep machine learning model

Aiming at the low accuracy of multimedia visual image cooperative classification, a new method of multimedia visual image cooperative classification based on depth machine learning model is proposed. Firstly, HSV color space model is selected to extract color features of multimedia visual images, Gabor function is used to extract texture features, and shape invariant moments are used to extract shape features. Then the features of multimedia visual images are recognized and classified, and the model parameters are optimized and adjusted according to the deviation of training output. The experimental results show that the average accuracy of this method is 95.892%, and the classification efficiency is high. The classification accuracy of this method is basically above 95%, and the classification accuracy is high. The training time of image type samples is 19 s, the testing time of image type is 12 s, and the time consumption is low.

The Emblem in the Visual Representation of a University: Classification and Semiotics of Visual Images

The article studies the main attribute of a university’s visual representation - its emblem. Among a few works devoted to the process of university branding, there is no сошріех analysis of their visual representation functioning. However, the observed growth of branding work in various universities increasingly involves the emblem as their graphic identifier. With the help of quantitative and qualitative methods, the structural and semiotic analysis of the emblems has been carried out on a соmріеіе sample of 917 Russian universities, their structure and range of their visual images being characterized, a typology of semiotic meanings being compiled. Consequently, the semiotic bases of Russian universities’ visual representation have been characterized, its functions in the process of university branding being defined. The research results have a wide practical application as a source of information on the use and purposes of university emblems within the higher education system, both for specialists involved in branding (marketing specialists, sociologists, designers etc.) and for PR departments, media centers, and administrations within a university. Nevertheless, the interpretative nature of the semiotic typology and the non-exclusive character of the classification of visual images allow but to designate the semantic directions of the universities positioning, beyond characterizing certain educational institutions in terms of their models of visual communication. Additional studies are necessary to determine the validity of the classification, as well as to clarify the semiotic bases for the typology given.