Showing papers on "Grayscale published in 2023"

Single-vat single-cure grayscale digital light processing 3D printing of materials with large property difference and high stretchability

Abstract: Multimaterial additive manufacturing has important applications in various emerging fields. However, it is very challenging due to material and printing technology limitations. Here, we present a resin design strategy that can be used for single-vat single-cure grayscale digital light processing (g-DLP) 3D printing where light intensity can locally control the conversion of monomers to form from a highly stretchable soft organogel to a stiff thermoset within in a single layer of printing. The high modulus contrast and high stretchability can be realized simultaneously in a monolithic structure at a high printing speed (z-direction height 1 mm/min). We further demonstrate that the capability can enable previously unachievable or hard-to-achieve 3D printed structures for biomimetic designs, inflatable soft robots and actuators, and soft stretchable electronics. This resin design strategy thus provides a material solution in multimaterial additive manufacture for a variety of emerging applications.

Complex aspherical singlet and doublet microoptics by grayscale 3D printing.

Abstract: We demonstrate 3D printed aspherical singlet and doublet microoptical components by grayscale lithography and characterize and evaluate their excellent shape accuracy and optical performance. The typical two-photon polymerization (2PP) 3D printing process creates steps in the structure which is undesired for optical surfaces. We utilize two-photon grayscale lithography (2GL) to create step-free lenses. To showcase the 2GL process, the focusing ability of a spherical and aspherical singlet lens are compared. The surface deviations of the aspherical lens are minimized by an iterative design process and no distinct steps can be measured via confocal microscopy. We design, print, and optimize an air-spaced doublet lens with a diameter of 300 µm. After optimization, the residual shape deviation is less than 100 nm for the top lens and 20 nm for the bottom lens of the doublet. We examine the optical performance with an USAF 1951 resolution test chart to find a resolution of 645 lp/mm.

Separable reversible data hiding in an encrypted image using the adjacency pixel difference histogram

Abstract: Information security is a practice of encrypting data in movement and on hold, improving discretion and integrity. One can protect, encrypt and decrypt critical data in several ways. One of them is reversible data hiding in an encrypted image. The technique enables the user to encrypt images that need authentication and restore them to their original form of images. This technique returns a lossless image as an output making it the most suiTable for medical images and the military. Histogram shifting of pixel difference is an effectual reversible data hiding method in information security. Each image's pixel is encrypted when a user wants to safely store a digital image in an open environment like a cloud. The authentication or any other relevant information related to that image is embedded in the pixel difference histogram. The proposed approach's advantage is that the grayscale image transfer is carried out exceedingly safely, with near-zero correlation and Entropy closer to 8. The Peak Signal Noise Ratio (PSNR) for a directly decrypted image with an embedding capacity of 0.0807 bpp is 50.84 dB. Moreover, the secret and cover images are retrieved without error.

Combining the YOLOv4 Deep Learning Model with UAV Imagery Processing Technology in the Extraction and Quantization of Cracks in Bridges

Abstract: Bridges are often at risk due to the effects of natural disasters, such as earthquakes and typhoons. Bridge inspection assessments normally focus on cracks. However, numerous concrete structures with cracked surfaces are highly elevated or over water, and is not easily accessible to a bridge inspector. Furthermore, poor lighting under bridges and a complex visual background can hinder inspectors in their identification and measurement of cracks. In this study, cracks on bridge surfaces were photographed using a UAV-mounted camera. A YOLOv4 deep learning model was used to train a model for identifying cracks; the model was then employed in object detection. To perform the quantitative crack test, the images with identified cracks were first converted to grayscale images and then to binary images the using local thresholding method. Next, the two edge detection methods, Canny and morphological edge detectors were applied to the binary images to extract the edges of the cracks and obtain two types of crack edge images. Then, two scale methods, the planar marker method, and the total station measurement method, were used to calculate the actual size of the crack edge image. The results indicated that the model had an accuracy of 92%, with width measurements as precise as 0.22 mm. The proposed approach can thus enable bridge inspections and obtain objective and quantitative data.

Color-Gray Multi-Image Hybrid Compression–Encryption Scheme Based on BP Neural Network and Knight Tour

Abstract: In the research of multi-image encryption (MIE), the image type and size are important factors that limit the algorithm design. For this reason, the multi-image (MI) hybrid encryption algorithm that can flexibly encrypt color images and grayscale images of various sizes is proposed. Based on this, combining the back propagation (BP) neural network compression technology and the MI hybrid encryption algorithm, an MI hybrid compression–encryption (MIHCE) scheme can be obtained to reduce the pressure of simultaneous transmission and storage of multiple cipher images. Besides, two chaotic maps are used in the scheme design process. By plotting the phase diagrams under different parameter conditions, the rich variation of the behavior of the chaotic maps in the phase space is exhibited. The MIHCE scheme based on the chaotic maps consists of three parts: 1) compressing the MI cube by using the BP neural network; 2) scrambling the compressed MI cube based on the knight tour problem and chaotic sequences; and 3) diffusing the scrambled MI cube. After the MIHCE is completed, the obtained cipher images are stored and transmitted. Subsequently, the security analysis and compression performance analysis prove the feasibility and safety of the designed compression–encryption scheme.

Pixel‐Level Grayscale Manipulation to Improve Accuracy in Digital Light Processing 3D Printing

Abstract: Digital light processing (DLP) is a widely used additive manufacturing technique for functional applications due to its high accuracy and print speeds. However, a variety of factors such as pixel size, motion stage resolution, optical focus, and chemical properties of the resin limit DLP's minimum resolution. Recently, research into locally varying light intensities has led to the emergence of grayscale DLP printing, which offers new capabilities including sub‐pixel manipulation of the printed shape. Here, a methodology is developed to enhance accuracy beyond what is typically capable for a given projector resolution by using pixel‐level grayscale control to create round features from sharp pixels. A numerical representation of the DLP pixel shape is developed to account for the effects of the incident light patterns. A reaction‐diffusion model is then used to predict the printed shapes before and after grayscale enhancement. This model is used to determine the optimal pixel intensities to match a target shape. Finally, the minimum feature size allowed by the proposed method is explored. The promising results represent an important step forward in raising DLP printing to higher accuracy, which will allow the fabrication of functional and structural components with smaller features or smoother faces.

Multi-Modal Stacking Ensemble for the Diagnosis of Cardiovascular Diseases

Abstract: Background: Cardiovascular diseases (CVDs) are a leading cause of death worldwide. Deep learning methods have been widely used in the field of medical image analysis and have shown promising results in the diagnosis of CVDs. Methods: Experiments were performed on 12-lead electrocardiogram (ECG) databases collected by Chapman University and Shaoxing People’s Hospital. The ECG signal of each lead was converted into a scalogram image and an ECG grayscale image and used to fine-tune the pretrained ResNet-50 model of each lead. The ResNet-50 model was used as a base learner for the stacking ensemble method. Logistic regression, support vector machine, random forest, and XGBoost were used as a meta learner by combining the predictions of the base learner. The study introduced a method called multi-modal stacking ensemble, which involves training a meta learner through a stacking ensemble that combines predictions from two modalities: scalogram images and ECG grayscale images. Results: The multi-modal stacking ensemble with a combination of ResNet-50 and logistic regression achieved an AUC of 0.995, an accuracy of 93.97%, a sensitivity of 0.940, a precision of 0.937, and an F1-score of 0.936, which are higher than those of LSTM, BiLSTM, individual base learners, simple averaging ensemble, and single-modal stacking ensemble methods. Conclusion: The proposed multi-modal stacking ensemble approach showed effectiveness for diagnosing CVDs.

Image encryption algorithm based on the density and 6D logistic map

Abstract: One of the most difficult issues in the history of communication technology is the transmission of secure images. On the internet, photos are used and shared by millions of individuals for both private and business reasons. Utilizing encryption methods to change the original image into an unintelligible or scrambled version is one way to achieve safe image transfer over the network. Cryptographic approaches based on chaotic logistic theory provide several new and promising options for developing secure Image encryption methods. The main aim of this paper is to build a secure system for encrypting gray and color images. The proposed system consists of two stages, the first stage is the encryption process, in which the keys are generated depending on the chaotic logistic with the image density to encrypt the gray and color images, and the second stage is the decryption, which is the opposite of the encryption process to obtain the original image. The proposed method has been tested on two standard gray and color images publicly available. The test results indicate to the highest value of peak signal-to-noise ratio (PSNR), unified average changing intensity (UACI), number of pixel change rate (NPCR) are 7.7268, 50.2011 and 100, respectively. While the encryption and decryption speed up to 0.6319 and 0.5305 second respectively.

Learning-based 3D imaging from single structured-light image

Abstract: Integrating structured-light technique with deep learning for single-shot 3D imaging has recently gained enormous attention due to its unprecedented robustness. This paper presents an innovative technique of supervised learning-based 3D imaging from a single grayscale structured-light image. The proposed approach uses a single-input, double-output convolutional neural network to transform a regular fringe-pattern image into two intermediate quantities which facilitate the subsequent 3D image reconstruction with high accuracy. A few experiments have been conducted to demonstrate the validity and robustness of the proposed technique.

A New Framework for Visual Classification of Multi-Channel Malware Based on Transfer Learning

Abstract: With the continuous development and popularization of the Internet, there has been an increasing number of network security problems appearing. Among them, the rapid growth in the number of malware and the emergence of variants have seriously affected the security of the Internet. Traditional malware detection methods require heavy feature engineering, which seriously affects the efficiency of detection. Existing deep-learning-based malware detection methods have problems such as poor generalization ability and long training time. Therefore, we propose a malware classification method based on transfer learning for multi-channel image vision features and ResNet convolutional neural networks. Firstly, the features of malware samples are extracted and converted into grayscale images of three different types. Then, the grayscale image sizes are processed using the bilinear interpolation algorithm to make them uniform in size. Finally, the three grayscale images are synthesized into three-dimensional RGB images, and the RGB images processed using data enhancement are used for training and classification. For the classification model, we used the previous ImageNet dataset (>10 million) and trained all the parameters of ResNet after loading the weights. For the evaluations, an experiment was conducted using the Microsoft BIG benchmark dataset. The experimental results showed that the accuracy on the Microsoft dataset reached 99.99%. We found that our proposed method can better extract the texture features of malware, effectively improve the accuracy and detection efficiency, and outperform the compared models on all performance metrics.

An Unpaired Thermal Infrared Image Translation Method Using GMA-CycleGAN

Abstract: Automatically translating chromaticity-free thermal infrared (TIR) images into realistic color visible (CV) images is of great significance for autonomous vehicles, emergency rescue, robot navigation, nighttime video surveillance, and many other fields. Most recent designs use end-to-end neural networks to translate TIR directly to CV; however, compared to these networks, TIR has low contrast and an unclear texture for CV translation. Thus, directly translating the TIR temperature value of only one channel to the RGB color value of three channels without adding additional constraints or semantic information does not handle the one-to-three mapping problem between different domains in a good way, causing the translated CV images not only to have blurred edges but also color confusion. As for the methodology of the work, considering that in the translation from TIR to CV the most important process is to map information from the temperature domain into the color domain, an improved CycleGAN (GMA-CycleGAN) is proposed in this work in order to translate TIR images to grayscale visible (GV) images. Although the two domains have different properties, the numerical mapping is one-to-one, which reduces the color confusion caused by one-to-three mapping when translating TIR to CV. Then, a GV-CV translation network is applied to obtain CV images. Since the process of decomposing GV images into CV images is carried out in the same domain, edge blurring can be avoided. To enhance the boundary gradient between the object (pedestrian and vehicle) and the background, a mask attention module based on the TIR temperature mask and the CV semantic mask is designed without increasing the network parameters, and it is added to the feature encoding and decoding convolution layers of the CycleGAN generator. Moreover, a perceptual loss term is applied to the original CycleGAN loss function to bring the translated images closer to the real images regarding the space feature. In order to verify the effectiveness of the proposed method, the FLIR dataset is used for experiments, and the obtained results show that, compared to the state-of-the-art model, the subjective quality of the translated CV images obtained by the proposed method is better, as the objective evaluation metric FID (Fréchet inception distance) is reduced by 2.42 and the PSNR (peak signal-to-noise ratio) is improved by 1.43.

Towards Homogeneous Modality Learning and Multi-Granularity Information Exploration for Visible-Infrared Person Re-Identification

Abstract: Visible-infrared person re-identification (VI-ReID) is a challenging and essential task that aims to retrieve a set of person images over visible and infrared camera views. In order to mitigate the impact of large modality discrepancy existing in heterogeneous images, previous methods attempt to apply generative adversarial network (GAN) to generate the modality-consisitent data. However, due to severe color variations between the visible domain and infrared domain, the generated fake cross-modality samples often fail to possess good qualities to fill the modality gap between synthesized scenarios and target real ones, which leads to sub-optimal feature representations. In this work, we address the cross-modality matching problem with Aligned Grayscale Modality (AGM), an unified dark-line spectrum that reformulates visible-infrared dual-mode learning as a gray-gray single-mode learning problem. Specifically, we generate the grayscale modality from the homogeneous visible images. Then, we train a style tranfer model to transfer infrared images into homogeneous grayscale images. In this way, the modality discrepancy is remarkably reduced in the image space. In order to reduce the remaining appearance discrepancy, we further introduce a multi-granularity feature extraction network to conduct feature-level alignment. Rather than relying on the global information, we propose to exploit local (head-shoulder) features to assist person Re-ID, which complements each other to form a stronger feature descriptor. Comprehensive experiments implemented on the mainstream evaluation datasets include SYSU-MM01 and RegDB indicate that our method can remarkably enhance cross-modality retrieval performance against the state of the art methods.

Grayscale Digital Light Processing Gradient Printing for Stress Concentration Reduction and Material Toughness Enhancement

Abstract: Avoiding stress concentrations is essential to achieve robust parts since failure tends to originate at such concentrations. With recent advances in multi-material additive manufacturing, it is possible to alter the stress (or strain) distribution by adjusting the material properties in select locations. Here we investigate the use of grayscale digital light processing (g-DLP) 3D printing to create modulus gradients around areas of high stress. These gradients prevent failure by redistributing high stresses (or strains) to neighboring materials. The improved material stiffness distributions are calculated using finite element analysis (FEA). The much-enhanced properties are demonstrated experimentally for thin plates with circular, triangular, and elliptical holes. This work suggests that multi-material additive manufacturing techniques like g-DLP provide a unique opportunity to create lighter, tougher engineering materials and parts.

Structure Representation Network and Uncertainty Feedback Learning for Dense Non-uniform Fog Removal

Abstract: Few existing image defogging or dehazing methods consider dense and non-uniform particle distributions, which usually happen in smoke, dust and fog. Dealing with these dense and/or non-uniform distributions can be intractable, since fog’s attenuation and airlight (or veiling effect) significantly weaken the background scene information in the input image. To address this problem, we introduce a structure-representation network with uncertainty feedback learning. Specifically, we extract the feature representations from a pre-trained Vision Transformer (DINO-ViT) module to recover the background information. To guide our network to focus on non-uniform fog areas, and then remove the fog accordingly, we introduce the uncertainty feedback learning, which produces uncertainty maps, that have higher uncertainty in denser fog regions, and can be regarded as an attention map that represents fog’s density and uneven distribution. Based on the uncertainty map, our feedback network refines our defogged output iteratively. Moreover, to handle the intractability of estimating the atmospheric light colors, we exploit the grayscale version of our input image, since it is less affected by varying light colors that are possibly present in the input image. The experimental results demonstrate the effectiveness of our method both quantitatively and qualitatively compared to the state-of-the-art methods in handling dense and non-uniform fog or smoke.

Reversible Data Hiding in Palette Images

Abstract: Reversible data hiding (RDH) has been investigated for over two decades. Depending on the application scenario, it can be divided into RDH and reversible data hiding in encrypted images (RDHEI). Interestingly, almost all studies on RDH/RDHEI have been conducted on gray-scale images, and relatively few studies have been conducted on color images compared to those on gray-scale images. Moreover, very few studies have been undertaken on palette images as a widely used image format. For palette images in which pixels are not gray-scale levels but color table indexes, it is difficult to apply traditional RDH/RDHEI methods directly. Therefore, we propose a new framework for RDH/RDHEI specifically for palette images. This framework adds a route selection algorithm before the conventional RDH/RDHEI approach. To be specific, we propose a method named the shortest route with correlation and frequency selection to reorder the color table, followed by a correlation reconstruction of the remapped index image according to this color table. The experimental results show the improvement brought by our proposed route selection to the existing RDH/RDHEI methods in palette images.

Substation Danger Sign Detection and Recognition using Convolutional Neural Networks

Abstract: This paper focuses on the training of a deep neural network regarding danger sign detection and recognition in a substation. It involved applying the concepts of neural networks and computer vision to achieve results similar to traffic sign and number plate detection systems. The input data were captured in three distinct formats, i.e. grayscale, RGB, and YCbCr, which have been used as a base for comparison in this paper. The efficiency of the neural network was tested on a unique data set involving danger signs present in industrial and processing facilities. The data set was unique, consisting of four distinct symbols. The trained data were selected so that they would not facilitate overfitting and also would not be under fitted. The accuracy of the model varied with the input type and was tested with two distinct classifiers, CNN and SVM, and the results were compared. The model was designed to be fast and accurate, and it can be implemented on mobile devices.

Dot profile model-based direct binary search

Abstract: A dot profile model to compensate dot shape irregularity errors of inkjet printers is proposed. Previous tabular approaches for parameterizing the printer model rely on the measurements of the gray level of various printed halftone patterns. However, lots of patterns need to be printed and scanned if the printer generates large drops of colorant. To solve this problem, we propose to simulate the appearance of the rendered patterns so that the model parameters can be computed analytically. The simulation uses the mean dot as the printer dot profile and saturated addition to resolve dot overlap. Besides, we incorporate a standard definition (SD) and a high definition (HD) equivalent gray-scale representation of the printed halftone image produced by the dot profile model into the direct binary search (DBS) algorithm. Experimental results show great improvement in the mid-tone and shadow regions over the printed image halftoned by the original DBS. The HD model further enhances details in the shadows.

Sensor Fusion Approach for Multiple Human Motion Detection for Indoor Surveillance Use-Case

Abstract: Multi-human detection and tracking in indoor surveillance is a challenging task due to various factors such as occlusions, illumination changes, and complex human-human and human-object interactions. In this study, we address these challenges by exploring the benefits of a low-level sensor fusion approach that combines grayscale and neuromorphic vision sensor (NVS) data. We first generate a custom dataset using an NVS camera in an indoor environment. We then conduct a comprehensive study by experimenting with different image features and deep learning networks, followed by a multi-input fusion strategy to optimize our experiments with respect to overfitting. Our primary goal is to determine the best input feature types for multi-human motion detection using statistical analysis. We find that there is a significant difference between the input features of optimized backbones, with the best strategy depending on the amount of available data. Specifically, under a low-data regime, event-based frames seem to be the preferred input feature type, while higher data availability benefits the combined use of grayscale and optical flow features. Our results demonstrate the potential of sensor fusion and deep learning techniques for multi-human tracking in indoor surveillance, although it is acknowledged that further studies are needed to confirm our findings.

Grayscale Image Statistical Attributes Effectively Distinguish the Severity of Lung Abnormalities in CT Scan Slices of COVID-19 Patients

Abstract: Grayscale statistical attributes analysed for 513 extract images taken from pulmonary computed tomography (CT) scan slices of 57 individuals (49 confirmed COVID-19 positive; eight confirmed COVID-19 negative) are able to accurately predict a visual score (VS from 0 to 4) used by a clinician to assess the severity of lung abnormalities in the patients. Some of these attributes can be used graphically to distinguish useful but overlapping distributions for the VS classes. Using machine and deep learning (ML/DL) algorithms with twelve grayscale image attributes as inputs enables the VS classes to be accurately distinguished. A convolutional neural network achieves this with better than 96% accuracy (only 18 images misclassified out of 513) on a supervised learning basis. Analysis of confusion matrices enables the VS prediction performance of ML/DL algorithms to be explored in detail. Those matrices demonstrate that the best performing ML/DL algorithms successfully distinguish between VS classes 0 and 1, which clinicians cannot readily do with the naked eye. Just five image grayscale attributes can also be used to generate an algorithmically defined scoring system (AS) that can also graphically distinguish the degree of pulmonary impacts in the dataset evaluated. The AS classification illustrated involves less overlap between its classes than the VS system and could be exploited as an automated expert system. The best-performing ML/DL models are able to predict the AS classes with better than 99% accuracy using twelve grayscale attributes as inputs. The decision tree and random forest algorithms accomplish that distinction with just one classification error in the 513 images tested.

Deep and Low-Rank Quaternion Priors for Color Image Processing

Abstract: Due to the physical nature of color images, color image processing such as denoising and inpainting has shown extensive and versatile possibilities over grayscale image processing. The monochromatic and the concatenation model have been widely used to process color images by processing each color channel independently or concatenating three color channels as one unified one and then used existing grayscale image processing methods directly without specific operations. These above schemes, however, have some limitations: (1) they would destroy the inherent correlation among three color channels since they cannot represent color images holistically; (2) they usually focus on one specific handcrafted prior such as smoothness, low-rankness, or even deep prior and thus failing to fuse deep and handcrafted priors of color images flexibly. To conquer these limitations, we propose one unified model to integrate deep prior and low-rank quaternion prior (DLRQP) for color image processing under the plug-and-play (PnP) framework. Specifically, the quaternion representation with low-rank constraint is introduced to denote the color image in a holistic way and one advanced denoiser is adopted to explore the deep prior in an iterative process. To tightly approximate the quaternion rank, one nonconvex penalty function is further utilized. We derive an alternate iterative approach to tackle the proposed model. We empirically demonstrate that our model can achieve superior performance over existing methods on both color image denoising and inpainting tasks.

SAR Edge Detection Using Weighted Directional Bhattacharyya Coefficients

Abstract: In this letter, an edge detector using weighted directional Bhattacharyya coefficients (WDBCs) is proposed to improve the edge detection performance on real synthetic aperture radar (SAR) images. The Bhattacharyya coefficient (BC) is a similarity measure that considers the grayscale distribution of pixels around edges and gives a better description of textured and nonuniform regions than the ratio of averages (ROAs). However, in addition to the grayscale distribution, the spatial distribution of pixels around the edges also has a significant impact on the edge strength. To consider both grayscale and spatial distribution, the weighted histograms are computed by Gaussian-Gamma-shaped (GGS) bi-windows to obtain WDBCs in eight directions. Then the edge strength map (ESM) and edge direction map (EDM) are defined by WDBCs. The edge detector can be obtained after non-maximum suppression and hysteresis thresholding. Experimental results on real SAR images compared with five existing detectors show the effectiveness of the proposed detector.