Neural Image Research Articles

Color imaging using a system based on 3 diffractive lenses

The possibility of essentially reducing the weight and production cost of computer vision systems has led to the publication of a large number of research works dealing with the development of new imaging systems based on diffractive optics. This study proposes a new imaging system composed of three diffractive lenses, with each forming a separate channel of the color RGB image. This approach allows us to significantly narrow the spectral range of each lens, thus significantly reducing the image distortion caused by chromatic aberration inherent in diffractive optics. It shows that this scheme allows us to perform the neural network-aided image reconstruction, providing a significantly improved resulting image quality. The study proposes a false edge level criterion (FEL) for evaluating the neural network-aided reconstruction.

Neural image caption generator based on crossbar array design of memristor module

The crossbar array design method based on a single memristor provides an excessive writing error and cannot achieve the network’s requirements for the generation of image captions. In order to solve this problem, this paper highlights the following techniques. (1) A crossbar array of memristor modules is built, and the writing error is theoretically reduced to a maximum of 0.0619‰. (2) A method of linearly mapping the weights of neural networks to memristance values is proposed, and a memristive neural network based on a memristor module is designed. (3) A neural image caption generator based on VGG-16-LSTM is designed in light of the crossbar array of the memristor module. The design can achieve a maximum of 99.38% of the reference network in BLEU scoring performance. The performance of neural networks based on memristor modules under different input conditions is discussed, and the paper’s best hardware design scheme, named it-10, is proposed. The it-8 simulation can obtain 97.37% of the reference network in BLEU scoring performance, and it requires far less demanding input conditions. Moreover, this paper finds a way to design a novel crossbar array based on the memristor module and make a breakthrough in memristive neural network solution for the cross-modal image caption generation task.

Picture This: A Deep Learning Model for Operational Real Estate Emissions

We present a deep learning model estimating carbon dioxide equivalent (CO2e) emissions in the real estate sector. The model, which utilizes convolutional neural networks (CNNs) and image classification techniques, is designed to estimate CO2e emissions based on publicly available images of buildings and their corresponding emissions. Our findings show that the model has the ability to provide reasonably accurate estimations of CO2e emissions using images as the sole input. Notably, incorporating primary energy sources as additional input further improves the accuracy up to 75%. The creation of such a model is particularly important in the fight against climate change, as it allows for transparency and fast identification of buildings, contributing significantly to CO2e emissions in the building sector. Currently, information on emission intensity in the real estate sector is scarce, with only a few countries collecting and providing the required data. Our model can help reduce this gap and provide valuable insights into the carbon footprint of the real estate sector.

Feature pyramid-based convolutional neural network image inpainting

Feature pyramid-based convolutional neural network image inpainting

A 4D discrete Hopfield neural network-based image encryption scheme with multiple diffusion modes

A novel 4D chaotic map is proposed in this paper using the model of a self-feedback Discrete Hopfield Neural Network (DHNN) with nonlinear synaptic ICMIC (4D-IDHNN). Based on an analysis of chaos characteristics, including attractors, Lyapunov exponents, Bifurcation, Sample Entropy and CO algorithms, the 4D-IDHNN system demonstrates reasonable stochasticity and ergodicity, alongside a broad spectrum of hyperchaotic parameters. Based on the 4D-IDHNN system, we design a new Pixel-dependent bit-plane Chaotic random Diffusion (PbCrD) algorithm with high diffusion capability and security. To begin, the 4D-IDHNN system generates pseudo-random sequences to confuse the plaintext image pixel positions with Chaotic Magic Random Scramble (CMRS). Subsequently, the PbCrD algorithm can be employed to diffuse pixel values. With the initial value of 4D-IDHNN generated using the SHA-256 hash of plaintext image, the ability to resist plaintext attacks can be enhanced. The algorithm includes 24 diffusion modes, which is three times greater than the number of DNA rules, thereby enhancing the complexity of decryption. The proposed encryption algorithm has good effectiveness, robustness, and resistance to differential attacks based on simulations and security analyses.

PolySeg Plus: Polyp Segmentation Using Deep Learning with Cost Effective Active Learning

A deep convolution neural network image segmentation model based on a cost-effective active learning mechanism is proposed and named PolySeg Plus. It is intended to address polyp segmentation with a lack of labeled data and a high false-positive rate of polyp discovery. In addition to applying active learning, which assisted in labeling more image samples, a comprehensive polyp dataset formed of five benchmark datasets was generated to increase the number of images. To enhance the captured image features, the locally shared feature method is used, which utilizes the power of employing neighboring features together with one another to improve the quality of image features and overcome the drawbacks of the Conditional Random Features method. Medical image segmentation was performed using ResUNet++, ResUNet, UNet++, and UNet models. Gaussian noise was removed from the images using a gaussian filter, and the images were then augmented before being fed into the models. In addition to optimizing model performance through hyperparameter tuning, grid search is used to select the optimum parameters to maximize model performance. The results demonstrated a significant improvement and applicability of the proposed method in polyp segmentation when compared to state-of-the-art methods on the datasets CVC-ClinicDB, CVC-ColonDB, ETIS Larib Polyp DB, KVASIR-SEG, and Kvasir-Sessile, with Dice coefficients of 0.9558, 0.8947, 0.7547, 0.9476, and 0.6023, respectively. Not only did the suggested method improve the dice coefficients on the individual datasets, but it also produced better results on the comprehensive dataset, which will contribute to the development of computer-aided diagnosis systems.

Artificial Intelligence's Use in the Diagnosis of Mouth Ulcers: A Systematic Review.

Artificial intelligence (AI) has been cited as being helpful in the diagnosis of diseases, the prediction of prognoses, and the development of patient-specific therapeutic strategies. AI can help dentists, in particular, when they need to make important judgments quickly. It can eliminate human mistakes in making decisions, resulting in superior and consistent medical treatment while lowering the workload on dentists.The existing studies relevant to the study and application of AI in the diagnosis of various forms of mouth ulcers are reviewed in this work. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) standards were followed in the preparation of the review. There were no rule violations, with the significant exception of the use of a better search method that led to more accurate findings. Using search terms mainly such as AI, oral health, oral ulcers, oral herpes simplex, oral lichen planus, pemphigus vulgaris, recurrent aphthous ulcer (RAU), oral cancer, premalignant and malignant disorders, etc., a comprehensive search was carried out in the reliable sources of literature, namely PubMed, Scopus, Embase, Web of Science, Ovid, Global Health, and PsycINFO. For all papers, exhaustive searches were done using inclusion criteria as well as exclusion criteria between June 28, 2018, and June 28, 2023. An AI framework for the automatic categorization of oral ulcers from oral clinical photographs was developed by the authors, and it performed satisfactorily. The newly designed AI model works better than the current convolutional neural network image categorization techniques and shows a fair level of precision in the classification of oral ulcers. However, despite being useful for identifying oral ulcers, the suggested technique needs a broader set of data for validation and training purposes before being used in clinical settings. Automated OCSCC identification using a deep learning-based technique is a quick, harmless, affordable, and practical approach to evaluating the effectiveness of cancer treatment. The categorization and identification of RAU lesions through the use of non-intrusive oral pictures using the previously developed ResNet50 and YOLOV algorithms demonstrated better accuracy as well as adequate potential for the future, which could be helpful in clinical practice. Moreover, the most reliable projections for the likelihood of the presence or absence of RAU were made by the optimized neural network. The authors also discovered variables associated with RAU that might be used as input information to build artificial neural networks that anticipate RAU.

Unknown cell class distinction via neural network based scattering snapshot recognition.

Neural network-based image classification is widely used in life science applications. However, it is essential to extrapolate a correct classification method for unknown images, where no prior knowledge can be utilised. Under a closed set assumption, unknown images will be inevitably misclassified, but this can be genuinely overcome choosing an open-set classification approach, which first generates an in-distribution of identified images to successively discriminate out-of-distribution images. The testing of such image classification for single cell applications in life science scenarios has yet to be done but could broaden our expertise in quantifying the influence of prediction uncertainty in deep learning. In this framework, we implemented the open-set concept on scattering snapshots of living cells to distinguish between unknown and known cell classes, targeting four different known monoblast cell classes and a single tumoral unknown monoblast cell line. We also investigated the influence on experimental sample errors and optimised neural network hyperparameters to obtain a high unknown cell class detection accuracy. We discovered that our open-set approach exhibits robustness against sample noise, a crucial aspect for its application in life science. Moreover, the presented open-set based neural network reveals measurement uncertainty out of the cell prediction, which can be applied to a wide range of single cell classifications.

Bone tumor examination based on FCNN-4s and CRF fine segmentation fusion algorithm

Background and objectiveBone tumor is a kind of harmful orthopedic disease, there are benign and malignant points. Aiming at the problem that the accuracy of the existing machine learning algorithm for bone tumor image segmentation is not high, a bone tumor image segmentation algorithm based on improved full convolutional neural network which consists fully convolutional neural network (FCNN-4s) and conditional random field (CRF). MethodologyThe improved fully convolutional neural network (FCNN-4s) was used to perform coarse segmentation on preprocessed images. Batch normalization layers were added after each convolutional layer to accelerate the convergence speed of network training and improve the accuracy of the trained model. Then, a fully connected conditional random field (CRF) was fused to refine the bone tumor boundary in the coarse segmentation results, achieving the fine segmentation effect. ResultsThe experimental results show that compared with the traditional convolutional neural network bone tumor image segmentation algorithm, the algorithm has a great improvement in segmentation accuracy and stability, the average Dice can reach 91.56%, the real-time performance is better. ConclusionCompared with the traditional convolutional neural network segmentation algorithm, the algorithm in this paper has a more refined structure, which can effectively solve the problem of over-segmentation and under-segmentation of bone tumors. The segmentation prediction has better real-time performance, strong stability, and can achieve higher segmentation accuracy.

Development of a localization-based algorithm for the prediction of leg ulcer etiology.

Diagnostic work-up of leg ulcers is time- and cost-intensive. This study aimed at evaluating ulcer location as a diagnostic criterium and providing a diagnostic algorithm to facilitate differential diagnosis. The study consisted of 277 patients with lower leg ulcers. The following five groups were defined: Venous leg ulcer, arterial ulcers, mixed ulcer, arteriolosclerosis, and vasculitis. Using computational surface rendering, predilection sites of different ulcer types were evaluated. The results were integrated in a multinomial logistic regression model to calculate the likelihood of a specific diagnosis depending on location, age, bilateral involvement, and ulcer count. Additionally, neural network image analysis was performed. The majority of venous ulcers extended to the medial malleolar region. Arterial ulcers were most frequently located on the dorsal aspect of the forefoot. Arteriolosclerotic ulcers were distinctly localized at the middle third of the lower leg. Vasculitic ulcers appeared to be randomly distributed and were markedly smaller, multilocular and bilateral. The multinomial logistic regression model showed an overall satisfactory performance with an estimated accuracy of 0.68 on unseen data. The presented algorithm based on ulcer location may serve as a basic tool to narrow down potential diagnoses and guide further diagnostic work-up.

Low-epsilon adversarial attack against a neural network online image stream classifier

An adversary intercepts a stream of images between a sender and a receiver neural network classifier. To minimize its footprint, the adversary only attacks a limited number of images within the stream. The adversary is interested in maximizing the number of successfully conducted attacks among all performed attacks. Upon the arrival of each image and before the arrival of the following image, the adversary must irrevocably decide whether it wants to attack the current image or not. The target model is a fixed deep neural network that may use any form of regularization. The adversary has query access to the target model, which can feed images and obtain the loss, which may contain regularization and classification loss terms. Since this paper’s proposed method needs classification loss term alone, it also suggests a novel method in which the adversary estimates the regularization loss term and eliminates it. All images are partitioned into three groups based on their after-attack classification loss and treated according to their group. Moreover, this paper provides some promising test results on various datasets.

Unsupervised neural network-based image restoration framework for pattern fidelity improvement and robust metrology

Unsupervised neural network-based image restoration framework for pattern fidelity improvement and robust metrology

Evaluation of Spiking Neural Nets-Based Image Classification Using the Runtime Simulator RAVSim.

Spiking Neural Networks (SNNs) help achieve brain-like efficiency and functionality by building neurons and synapses that mimic the human brain's transmission of electrical signals. However, optimal SNN implementation requires a precise balance of parametric values. To design such ubiquitous neural networks, a graphical tool for visualizing, analyzing, and explaining the internal behavior of spikes is crucial. Although some popular SNN simulators are available, these tools do not allow users to interact with the neural network during simulation. To this end, we have introduced the first runtime interactive simulator, called Runtime Analyzing and Visualization Simulator (RAVSim),a developed to analyze and dynamically visualize the behavior of SNNs, allowing end-users to interact, observe output concentration reactions, and make changes directly during the simulation. In this paper, we present RAVSim with the current implementation of runtime interaction using the LIF neural model with different connectivity schemes, an image classification model using SNNs, and a dataset creation feature. Our main objective is to primarily investigate binary classification using SNNs with RGB images. We created a feed-forward network using the LIF neural model for an image classification algorithm and evaluated it by using RAVSim. The algorithm classifies faces with and without masks, achieving an accuracy of 91.8% using 1000 neurons in a hidden layer, 0.0758 MSE, and an execution time of ∼10[Formula: see text]min on the CPU.The experimental results show that using RAVSim not only increases network design speed but also accelerates user learning capability.

Design of high performance and energy efficient convolution array for convolution neural network-based image inference engine

The energy efficiency of CNN-based inference engines predominately depends upon Giga-operations-per-second and power consumption. The sparse-based accelerator compresses the insignificant inputs (input feature maps & weights), skips the inefficient computations, and improves energy efficiency. A sparse accelerator for weights could impact the accuracy of the inference. Therefore, a sparse network for Input Feature Maps (IFMs) is considered. MATLAB-based sparsity analysis is done layer-wise on the pre-trained CNN models like AlexNet, VGG-16, VGG-19, ResNet-18 & ResNet-34. Layer-wise analysis reveals that ∼18%–90% of the IFMs are zeros. Besides, IFMs and Weights adopted the 16-bit Fix/Float data format to maintain an accuracy as close as 97% with Single Precision Floating Point (SPFP). A 3 × 1 Convolutional array with improved Zero-detect-Skip (CZS3×1) control units for multiplier and adder/subtractor arrays is proposed. The modified rectified linear unit (RELU) converts IFM values ≤ 0 to zero and sets Detection-Bit (DB) to 1. These DBs decide the mode of effective zero-skip operations in CZS3×1. A 3 × 3 Compressed Processing Element (CPE) is designed using three CZS3×1. The 20-CPEs convolution array architecture is implemented in 65 nm technology libraries. The performance of 90 Giga-operations per second (GOP/s) and energy efficiency of 3.42 Tera operations per second per watt (TOPS/W) were attained for the proposed CPE. The CPE with improved control strategy enhanced the performance by a factor of 2.45 while consuming 8.8 times less energy on average than the state-of-the-art CNN accelerators.

Rapid appearance quality of rice based on machine vision and convolutional neural network research on automatic detection system.

In the process of rice production and storage, there are many defects in the traditional detection methods of rice appearance quality, but using modern high-precision instruments to detect the appearance quality of rice has gradually developed into a new research trend at home and abroad with the development of agricultural artificial intelligence. In this study, we independently designed a fast automatic rice appearance quality detection system based on machine vision technology by introducing convolutional neural network and image processing technology. In this study, NIR and RGB images were generated into five-channel image data by superposition function, and image are preprocessed by combining the Watershed algorithm with the Otus adaptive threshold function. Different grains in the samples were labeled and put in the convolutional neural network for training. The rice grains were classified and the phenotype data were analyzed by selecting the optimal training model to realize the detection of rice appearance quality. The experimental results showed that the resolution of the system could reach 92.3%. In the detection process, the system designed with this method not only reduces the subjectivity problems caused by different detection environments, visual fatigue caused large sample size and the inspector's personal factors, but also significantly improves the detection time and accuracy, which further enhances the detection efficiency of rice appearance quality, and has positive significance for the development of the rice industry.

Detection of extracranial and intracranial calcified carotid artery atheromas in cone beam computed tomography using a deep learning convolutional neural network image segmentation approach

We leveraged an artificial intelligence deep-learning convolutional neural network (DL CNN) to detect calcified carotid artery atheromas (CCAAs) on cone beam computed tomography (CBCT) images. We obtained 137 full-volume CBCT scans with previously diagnosed CCAAs. The DL model was trained on 170 single axial CBCT slices, 90 with extracranial CCAAs and 80 with intracranial CCAAs. A board-certified oral and maxillofacial radiologist confirmed the presence of each CCAA. Transfer learning through a U-Net-based CNN architecture was utilized. Data allocation was 60% training, 10% validation, and 30% testing. We determined the accuracy of the DL model in detecting CCAA by calculating the mean training and validation accuracy and the area under the receiver operating characteristic curve (AUC). We reserved 5 randomly selected unseen full CBCT volumes for final testing. The mean training and validation accuracy of the model in detecting extracranial CCAAs was 92% and 82%, respectively, and the AUC was 0.84 with 1.0 sensitivity and 0.69 specificity. The mean training and validation accuracy in detecting intracranial CCAAs was 61% and 70%, respectively, and the AUC was 0.5 with 0.93 sensitivity and 0.08 specificity. Testing of full-volume scans yielded an AUC of 0.72 and 0.55 for extracranial and intracranial CCAAs, respectively. Our DL model showed excellent discrimination in detecting extracranial CCAAs on axial CBCT images and acceptable discrimination on full-volumes but poor discrimination in detecting intracranial CCAAs, for which further research is required.

Research on Printmaking Image Classification and Creation Based on Convolutional Neural Network

As an important form of expression in modern civilization art, printmaking has a rich variety of types and a prominent sense of artistic hierarchy. Therefore, printmaking is highly favored around the world due to its unique artistic characteristics. Classifying print types through image feature elements will improve people’s understanding of print creation. Convolutional neural networks (CNNs) have good application effects in the field of image classification, so CNN is used for printmaking analysis. Considering that the classification effect of the traditional convolutional neural image classification model is easily affected by the activation function, the T-ReLU activation function is introduced. By utilizing adjustable parameters to enhance the soft saturation characteristics of the model and avoid gradient vanishing, a T-ReLU convolutional model is constructed. A better convolutional image classification model is proposed based on the T-ReLU convolutional model, taking into account the issue of subpar multi-level feature fusion in deep convolutional image classification models. Utilize normalization to analyze visual input, an eleven-layer convolutional network with residual units in the convolutional layer, and cascading thinking to fuse convolutional network defects. The performance test results showed that in the data test of different styles of artificial prints, the GT-ReLU model can obtain the best image classification accuracy, and the image classification accuracy rate is 0.978. The GT-ReLU model maintains a classification accuracy above 94.4% in the multi-dataset test classification performance test, which is higher than that of other image classification models. For the use of visual processing technology in the field of classifying prints, the research content provides good reference value.

Analysis of Convolutional Neural Network Image Classifiers in a Rotationally Symmetric Model

Convolutional neural network image classifiers are defined and the rate of convergence of the misclassification risk of the estimates towards the optimal misclassification risk is analyzed. Here we consider images as random variables with values in some functional space, where we only observe discrete samples as function values on some finite grid. Under suitable structural and smoothness assumptions on the functional a posteriori probability, which includes some kind of symmetry against rotation of subparts of the input image, it is shown that least squares plug-in classifiers based on convolutional neural networks are able to circumvent the curse of dimensionality in binary image classification if we neglect a resolution-dependent error term. The finite sample size behavior of the classifier is analyzed by applying it to simulated and real data.

Image enhancement based on improved pulse coupled neural network

Image is the main source of information acquisition and exchange. Image processing technology can realize man-machine dialogue, but it still faces many challenges, such as the lack of image edge information. In order to better realize the neural network image processing, this paper builds two kinds of neural network models based on memristor on the MATLAB platform. Then these two neural network models are used to enhance the image. The image is illustrated by the comparison of the histogram, the enhanced image and some objective criteria through the pulse coupled neural network and the improved neural network. The improved neural network is more efficient than the pulse coupled neural network in processing the edges and internal details of the image, highlighting the outline of the key part of the image, and effectively realizing the image enhancement. The advantages of the neural network with memristor in image processing are illustrated.

Deep Learning-assisted Retinopathy of Prematurity (ROP) Screening

Retinopathy of prematurity (ROP) is a leading cause of blindness in premature infants worldwide, particularly in developing countries. In this research, we propose a Deep Convolutional Neural Network (DCNN) and image processing-based approach for the automatic detection of retinal features, including the optical disc (OD) and retinal blood vessels (BV), as well as disease classification using a rule-based method for ROP patients. Our DCNN model uses YOLO-v5 for OD detection and either Pix2Pix or a U-Net for BV segmentation. We trained our DCNN models on publicly available fundus image datasets of size 1,117 and 288 for OD detection and BV segmentation, respectively. We evaluated our approach on a dataset of 439 preterm neonatal retinal images, testing for ROP Zone and 6 BV masks. Our proposed system achieved excellent results, with the OD detection module achieving an overall accuracy of 98.94% (when IoU 0.5) and the BV segmentation module achieving an accuracy of 96.69% and a Dice coefficient between 0.60 and 0.64. Moreover, our system accurately diagnosed ROP in Zone-1 with 88.23% accuracy. Our approach offers a promising solution for accurate ROP screening and diagnosis, particularly in low-resource settings, where it has the potential to improve healthcare outcomes.