Image Converter Research Articles

A study on the application of neural style transformation technique in personalized art creation

Abstract In this paper, using the loss function of image stylization, combined with the development of image stylization based on deep learning, the proposed convolutional neural network in the style of image conversion fitting and overfitting response. Based on this, the multi-scale feature fusion method is chosen to train the style conversion network, with the help of deep feature extraction of the image for style conversion, reconstruct the multi-scale feature fusion image, and send it to the decoder for deep coding to realize the style conversion. To evaluate the effectiveness of the proposed multi-scale feature fusion style conversion algorithm, the content loss and style loss parameters of the algorithm are analyzed using the lightweight encoder and VGG encoder, respectively. Calculate the number of algorithmic model parameters and the amount of computation. Analyze the change in the iteration number of the personalized art style conversion process and select the performance evaluation index to evaluate the results of the personalized art style image conversion process. In the process of personalized art style conversion, the multi-scale feature fusion algorithm network proposed in this paper can basically reconstruct the original image after 1000 iterations. Personalized style image reconstruction has a PSNR of 25.26 dB when the number of iterations is 1000. With the deepening of training, the reconstruction effect becomes better and better, and the advantages of personalized art style image conversion applications are significant.

Enhancing architectural image processing: A novel 2D to 3D algorithm using improved convolutional neural networks

In light of the escalating advancements in architectural intelligence and information technology, the construction of smart cities increasingly necessitates a higher degree of precision in architectural measurements. Conventional approaches to architectural measurement, characterized by their low efficiency and protracted execution time, need to be revised to meet these burgeoning demands. To address this gap, we introduce a novel architectural image processing model that synergistically integrates Restricted Boltzmann Machines (RBMs) with Convolutional Neural Networks (CNNs) to facilitate the conversion of 2D architectural images into 3D. In the implementation phase of the model, an initial preprocessing of the architectural images is performed, followed by depth map conversion via bilateral filtering. Subsequently, minor voids in the images are rectified through a neighborhood interpolation algorithm. Finally, the preprocessed 2D images are input into the integrated model of RBMs and CNNs, realizing the 2D to 3D conversion. Experimental outcomes substantiate that this novel model attains a precision rate of 97%, and significantly outperforms comparative algorithms in terms of both runtime and efficiency. These results compellingly corroborate our model?s superiority in architectural image processing, enhancing measurement accuracy and drastically reducing execution time.

ONDL: An optimized Neutrosophic Deep Learning model for classifying waste for sustainability.

Sustainability has become a key factor on our planet. If this concept is applied correctly, our planet will be greener and more eco-friendly. Nowadays, waste classification and management practices have become more evident than ever. It plays a crucial role in the sustainability ecosystem. Computer algorithms and deep learning can help in this sustainability challenge. In this paper, An Optimized Neutrosophic Deep Learning (ONDL) model was proposed to classify waste objects. Two datasets were tested in this research {Dataset for Waste Management 1 (DSWM1), and Dataset for Waste Management 2 (DSWM2)}. DSWM1 consists of two classes (Organic or Recycled) objects. The DSWM2 consists of three classes (Organic, Recycled, or Non-Recyclable) objects. Both datasets exist publicly on the internet. The ONDL model architecture is constructed based on Alexnet as a Deep Transfer Learning (DTL) model and the conversion of images to True (T) neutrosophic domain and Grey Wolf Optimization (GWO) for the image features selection. The selection process of the building components of the ONDL model is comprehensive as different DTL models (Alexnet, Googlenet, and Resnet18) are tested, and three neutrosophic domains (T, I, and F) domain are included. The ONDL model proved its efficiency against all the tested models, moreover, it achieves competitive results with related works in terms of testing accuracy and performance metrics. In DSWM1, the ONDL model achieved 0.9189, 0.9177, 0.9176, and 0.9177 in Testing Accuracy (TA), Precision (P), Recall (R), and F1 score. In DSWM2, it achieved 0.8532, 0.7728, 0.7944, and 0.7835 in TA, P, R, and F1 Score consequently.

Leveraging deep learning and image conversion of executable files for effective malware detection: A static malware analysis approach

<abstract><p>The escalating sophistication of malware poses a formidable security challenge, as it evades traditional protective measures. Static analysis, an initial step in malware investigation, involves code scrutiny without actual execution. One static analysis approach employs the conversion of executable files into image representations, harnessing the potency of deep learning models. Convolutional neural networks (CNNs), particularly adept at image classification, have potential for malware detection. However, their inclination towards structured data requires a preprocessing phase to convert software into image-like formats. This paper outlines a methodology for malware detection that involves applying deep learning models to image-converted executable files. Experimental evaluations have been performed by using CNN models, autoencoder-based models, and pre-trained counterparts, all of which have exhibited commendable performance. Consequently, employing deep learning for image-converted executable analysis emerges as a fitting strategy for the static analysis of software. This research is significant because it utilized the largest dataset to date and encompassed a wide range of deep learning models, many of which have not previously been tested together.</p></abstract>

Quantitative analysis of the effect of illumination variations on image processing in machine vision inspection

Abstract When unavoidable light, age, expression, and gesture changes occur, the machine vision detection accuracy will be greatly reduced, especially the light change impact on image processing. Combining the reflection equation and frequency domain analysis to construct a light change model based on spherical harmonic function and process the image from four directions, namely, digital image conversion, median filtering, sharpening processing, and image global segmentation. Aiming at the shortcomings of traditional image processing algorithms, the light compensation based on the improved pulse-coupled neural network model is proposed, and the model is used to analyze the impact of light changes on image processing. Despite the changes in light conditions, this algorithm maintains above 0.95 in each subset and has the best overall performance, as shown by the results. When the light intensity is less than 41.88kLux, using the optimal threshold for background segmentation of adult green peppers with respect to a fixed threshold T=0.3 results in a significant reduction of the background segmentation error. When the light intensity is greater than 41.88kLux, the superiority of the optimal threshold is not obvious. The results of this research not only have a significant role in promoting the development of the field of computer vision but also can provide people with convenient and efficient services in all aspects of social life.

Cascaded Thinning in Upscale and Downscale Representation for EEG Signal Processing.

Smoothing filters are widely used in EEG signal processing for noise removal while preserving signals' features. Inspired by our recent work on Upscale and Downscale Representation (UDR), this paper proposes a cascade arrangement of some effective image-processing techniques for signal filtering in the image domain. The UDR concept is to visualize EEG signals at an appropriate line width and convert it to a binary image. The smoothing process is then conducted by skeletonizing the signal object to a unit width and projecting it back to the time domain. Two successive UDRs could result in a better-smoothing performance, but their binary image conversion should be restricted. The process is computationally ineffective, especially at higher line width values. Cascaded Thinning UDR (CTUDR) is proposed, exploiting morphological operations to perform a two-stage upscale and downscale within one binary image representation. CTUDR is verified on a signal smoothing and classification task and compared with conventional techniques, such as the Moving Average, the Binomial, the Median, and the Savitzky Golay filters. Simulated EEG data with added white Gaussian noise is employed in the former, while cognitive conflict data obtained from a 3D object selection task is utilized in the latter. CTUDR outperforms its counterparts, scoring the best fitting error and correlation coefficient in signal smoothing while achieving the highest gain in Accuracy (0.7640%) and F-measure (0.7607%) when used as a smoothing filter for training data of EEGNet.

Biomedical Signal to Image Conversion and Classification using Flexible Deep Learning Techniques

Biomedical Signal to Image Conversion and Classification using Flexible Deep Learning Techniques

Signal to Image Conversion and Convolutional Neural Networks for Physiological Signal Processing: A Review

Signal to Image Conversion and Convolutional Neural Networks for Physiological Signal Processing: A Review

Dual-3DM3AD: Mixed Transformer Based Semantic Segmentation and Triplet Pre-Processing for Early Multi-Class Alzheimer's Diagnosis.

Alzheimer's Disease (AD) is a widespread, chronic, irreversible, and degenerative condition, and its early detection during the prodromal stage is of utmost importance. Typically, AD studies rely on single data modalities, such as MRI or PET, for making predictions. Nevertheless, combining metabolic and structural data can offer a comprehensive perspective on AD staging analysis. To address this goal, this paper introduces an innovative multi-modal fusion-based approach named as Dual-3DM3-AD. This model is proposed for an accurate and early Alzheimer's diagnosis by considering both MRI and PET image scans. Initially, we pre-process both images in terms of noise reduction, skull stripping and 3D image conversion using Quaternion Non-local Means Denoising Algorithm (QNLM), Morphology function and Block Divider Model (BDM), respectively, which enhances the image quality. Furthermore, we have adapted Mixed-transformer with Furthered U-Net for performing semantic segmentation and minimizing complexity. Dual-3DM3-AD model is consisted of multi-scale feature extraction module for extracting appropriate features from both segmented images. The extracted features are then aggregated using Densely Connected Feature Aggregator Module (DCFAM) to utilize both features. Finally, a multi-head attention mechanism is adapted for feature dimensionality reduction, and then the softmax layer is applied for multi-class Alzheimer's diagnosis. The proposed Dual-3DM3-AD model is compared with several baseline approaches with the help of several performance metrics. The final results unveil that the proposed work achieves 98% of accuracy, 97.8% of sensitivity, 97.5% of specificity, 98.2% of f-measure, and better ROC curves, which outperforms other existing models in multi-class Alzheimer's diagnosis.

Biomedical signal to image conversion and classification using flexible deep learning techniques

Biomedical signal to image conversion and classification using flexible deep learning techniques

An Efficient VLSI Architecture for Bi-Cubic Interpolation using Carry Skip Adder

In any signal processing system, signal distortion can occur. A compensation scheme would adjust the captured signal to correct for any deviations and ensure accurate signal transmission and preservation. A compensation scheme is used to correct such deviations for mostly linear is used to enhance the quality of reconstructed signal suitable for high-quality super-resolution techniques. This technique is applicable for multidimensional signal processing such as image processing. Except for the resolution of the acquired images, the extracting of colour information is greatly affected by the number of pixels. This scheme enhances the correlation of interpolated pixels and their immediate neighbouring pixels. This research aims to develop a super-resolution demosaicking (SRD) technique to reconstruct high-resolution, full-colour images from samples of the same previously seen (in-painted) image without a sophisticated, time-consuming training process. Hardware-oriented colour demosaicking methods, which give importance to green colour pixels, have thus enhanced the quality of the restored image. A Boundary Mirror and Detector mechanism also considered here for improving the pixel quality up to the boundaries. Three interpolators were economized by using a method called hardware sharing. The present Bi-Cubic interpolation technique has been designed as part of our overall effort to enhance and advance the SRD approach toward better performance and broader acceptance. This had been implemented by VLSI architecture and VHDL, whereas MATLAB had executed the image conversion and de-conversion methods. Comparison in the evaluation of results was done based on PSNR and SSIM values, followed by the analysis of the output results after the synthesis process using a Xilinx Vivado ZYNQ-7-ZC702 FPGA.

2D to 3D Image Conversion Algorithms

2D to 3D Image Conversion Algorithms

High-Performance Encryption Algorithms for Dynamic Images Transmission

With the proliferation of the internet, the issue of image tampering has escalated, necessitating robust image encryption schemes. Despite the multitude of proposed image encryption algorithms, challenges such as slow computational speed, weak security, and suboptimal visibility persist. This study addresses these challenges by introducing a high-performance encryption algorithm tailored for dynamic images—QEDI (Quick Encryption Algorithm for Dynamic Images). QEDI leverages elliptic curve-based asymmetric encryption algorithms and hash algorithms. The process involves using a hash algorithm to calculate the hash of the dynamic image slated for encryption, employing the elliptic curve algorithm to generate public and private keys, and utilizing the asymmetric encryption algorithm to process the image hash, resulting in a signature ciphertext. This ciphertext is embedded into the dynamic image, completing the encryption process. To enhance the execution speed without compromising the image quality and visibility, the bit depth of the dynamic image is reduced before hash calculation, and custom information fields are encapsulated to embed the signature ciphertext into the dynamic image. Experimental evaluations, conducted within a dynamic image generation system, encompassed assessments of security, encryption algorithm execution time, and image quality. The results indicate that utilizing QEDI for dynamic image encryption yields better security compared to EDI (Encryption Algorithm for Dynamic Images). QEDI exhibits minimal impact on image quality, with a noteworthy 69.77% reduction in execution time compared to EDI. Furthermore, when compared to existing algorithms employing image conversion for encryption, QEDI-encrypted images demonstrate enhanced visibility, better image quality, and expedited encryption speed.

Generative adversarial networks for rubber stamp extraction and removal

Due to the variety of documents and people's different stamping habits, it is difficult for current models to achieve satisfactory results in the process of extracting and removing rubber stamps. Most current methods cannot extract or remove rubber stamps without destroying the background text information. To solve the above problems, we propose a neural network that uses a double-cycle structure, named stamp extraction and removal generative adversarial network (SERGAN). In detail, we improve SERGAN's cycle-consistency by simultaneously recombining the generated rubber stamp image and unstamped image into a stamped image, which enables SERGAN to have multi-domain image conversion capability. To prove the model's generalization ability, we compare it with current mainstream methods on the public watermark dataset CLWD, and the experimental results show the excellence of our model. In addition, there is no publicly available dataset for rubber stamp extraction and removal experiments. To address the scarcity of rubber stamp datasets, we construct a large-scale rubber stamp dataset named LRSD.

Light-weight color image conversion like pencil drawing for high-level synthesized hardware

Light-weight color image conversion like pencil drawing for high-level synthesized hardware

Artificial intelligence-aided diagnostic imaging: A state-of-the-art technique in precancerous screening.

Chromoendoscopy with the use of indigo carmine (IC) dye is a crucial endoscopic technique to identify gastrointestinal neoplasms. However, its performance is limited by the endoscopist's skill, and no standards are available for lesion identification. Thus, we developed an artificial intelligence (AI) model to replace chromoendoscopy. This pilot study assessed the feasibility of our novel AI model in the conversion of white-light images (WLI) into virtual IC-dyed images based on a generative adversarial network. The predictions of our AI model were evaluated against the assessments of five endoscopic experts who were blinded to the purpose of this study with a staining quality rating from 1 (unacceptable) to 4 (excellent). The AI model successfully transformed the WLI of polyps with different morphologies and different types of lesions in the gastrointestinal tract into virtual IC-dyed images. The quality ratings of the real IC-dyed and AI images did not significantly differ concerning surface structure (AI vs IC: 3.08 vs 3.00), lesion border (3.04 vs 2.98), and overall contrast (3.14 vs 3.02) from 10 sets of images (10 AI images and 10 real IC-dyed images). Although the score depended significantly on the evaluator, the staining methods (AI or real IC) and evaluators had no significant interaction (P>0.05) with each other. Our results demonstrated the feasibility of employing AI model's virtual IC staining, increasing the possibility of being employed in daily practice. This novel technology may facilitate gastrointestinal lesion identification in the future.

Machine learning based fileless malware traffic classification using image visualization

In today’s interconnected world, network traffic is replete with adversarial attacks. As technology evolves, these attacks are also becoming increasingly sophisticated, making them even harder to detect. Fortunately, artificial intelligence (AI) and, specifically machine learning (ML), have shown great success in fast and accurate detection, classification, and even analysis of such threats. Accordingly, there is a growing body of literature addressing how subfields of AI/ML (e.g., natural language processing (NLP)) are getting leveraged to accurately detect evasive malicious patterns in network traffic. In this paper, we delve into the current advancements in ML-based network traffic classification using image visualization. Through a rigorous experimental methodology, we first explore the process of network traffic to image conversion. Subsequently, we investigate how machine learning techniques can effectively leverage image visualization to accurately classify evasive malicious traces within network traffic. Through the utilization of production-level tools and utilities in realistic experiments, our proposed solution achieves an impressive accuracy rate of 99.48% in detecting fileless malware, which is widely regarded as one of the most elusive classes of malicious software.

Automatic Segmentation of Histological Images of Mouse Brains

Using a high-throughput neuroanatomical screen of histological brain sections developed in collaboration with the International Mouse Phenotyping Consortium, we previously reported a list of 198 genes whose inactivation leads to neuroanatomical phenotypes. To achieve this milestone, tens of thousands of hours of manual image segmentation were necessary. The present work involved developing a full pipeline to automate the application of deep learning methods for the automated segmentation of 24 anatomical regions used in the aforementioned screen. The dataset includes 2000 annotated parasagittal slides (24,000 × 14,000 pixels). Our approach consists of three main parts: the conversion of images (.ROI to .PNG), the training of the deep learning approach on the compressed images (512 × 256 and 2048 × 1024 pixels of the deep learning approach) to extract the regions of interest using either the U-Net or Attention U-Net architectures, and finally the transformation of the identified regions (.PNG to .ROI), enabling visualization and editing within the Fiji/ImageJ 1.54 software environment. With an image resolution of 2048 × 1024, the Attention U-Net provided the best results with an overall Dice Similarity Coefficient (DSC) of 0.90 ± 0.01 for all 24 regions. Using one command line, the end-user is now able to pre-analyze images automatically, then runs the existing analytical pipeline made of ImageJ macros to validate the automatically generated regions of interest resulting. Even for regions with low DSC, expert neuroanatomists rarely correct the results. We estimate a time savings of 6 to 10 times.

A novel code generator for graphical user interfaces

Graphical user interfaces (GUIs) are widely used in human–computer interaction, providing a convenient interface for operation. Automating the conversion of GUI design images into source code can significantly reduce the coding workload for front-end developers. Detecting elements in GUI images is a key challenge in achieving automatic GUI code generation and is crucial for tasks such as GUI automation and testing. However, current state-of-the-art methods do not fully consider the unique characteristics of GUI images and elements, and they lack the required high localization accuracy, resulting in low detection accuracy for GUI element boxes. In this paper, we propose GUICG, an automatic GUI code generator that combines deep neural networks with image processing techniques to efficiently detect GUI elements from GUI images and generate front-end code. We empirically investigate various deep learning approaches and image processing methods for GUI component detection. Based on a comprehensive understanding of their performance and characteristics, we design GUICG by fusing image processing with a deep learning-based target detection model, achieving state-of-the-art performance. GUICG outperforms existing methods in accuracy and F1 score for component detection tasks, while producing human-readable code with a logical structure. Furthermore, we conduct an ablation study to quantitatively assess the impact of each key element in GUICG.

Abstract 17760: Conversion of Dicom ECG Images to Tabular Format for Building Large Language Model in Diagnoses and Disease Progression of Cardiovascular Conditions

Introduction: DICOM Electrocardiography (ECG) images from individuals are routinely stored at the institutional PACS server, including normal and abnormal findings. Hypothesis: The study’s objective is to create a database of ECGs as per Standard Communication Protocol (SCP) and to build an accurate Large Language Model (LLM) across different cardiovascular conditions Methods: DICOM ECGs are retrieved, anonymized and labelled to convert the signals to (x, y) coordinates with the help of PyDicom libraries. The tabular format is then stored with lead definition and dimensions of duration and amplitude, metadata, clinical conditions, and textual diagnoses. Results: In the pilot study, 1308 ECGs from individuals with different age (<45 & >45 years), gender (male & female), binary clinical categories (normal & abnormal) and heart rate (<90 & >90 per min) are selected for analysis and modeling. The DICOM image signal reproducibility following conversion (Figure 1 & 2) shows Cross Correlation Percentage of 0.85 (0.76 - 0.94) (Figure 3). An eXtreme Gradient Boosting (XGBoost) model (6) was trained to predict above variables. The results (Figure 4) show AUC highest for predicting binary clinical categories (normal vs abnormal) at 0.81 (0.7 - 0.9) and lowest for predicting age (<45 & >45 years) at 0.6 (0.46 - 0.7). The accuracies in the leads also show similar trend. The limitations include reproducibility of results in Lead VIII, IX and XI, which are sub average and hence being retrained with optimizers. These are initial results with relatively smaller database where multiclass disease or condition classifications are not performed. Conclusions: To conclude, the methodology possesses opportunities to improve the model with Deep Learning and Entity Disambiguation Techniques (NLP) to build Large Language ECG Models for cardiovascular diagnoses and disease progression trajectories.