Layout Image Research Articles

A Typology of Minimal Homepage Design: The Case of Global Fashion Brands

Fashion brands appeal to consumers based primarily on their design. Fashion brand websites act as avenues of communication with a strong appeal on consumers’ online behavior. At the same time, their minimal design signals high quality and aesthetics. Fashion brands’ website design has attracted academic attention, yet there exists no study on specific design dimensions of minimal websites in the fashion sector. A content analysis on 198 fashion brand websites demonstrates that negative space, homogenous layout, and colorful images prevail in the minimal websites of fashion brands, while text is restricted. Further, in-depth interviews with five graphic designers provide evidence with respect to their perceptions of minimal website design, its signals and appropriateness in the case of fashion brands.

Research on the optimization of interior design visual elements and art design innovation method under the deep convolutional network framework

Abstract As data science and computer science continue to advance, the application of deep learning methods to address technological innovation and optimize interior design has emerged as a significant area of research. In this paper, we propose a VR matching algorithm that utilizes deep convolutional networks to match the appropriate visual elements of the room layout, thereby realizing a panoramic scheme for interior design. A method of automatic extraction of features and labels of room layout based on a convolutional neural network is designed, trained, and tested with the help of a convolutional deep confidence network, and the labeled room layout feature map is classified using the Softmax classifier. A fast retrieval algorithm is being investigated with input in the form of images and combining text and content. Firstly, we input the room layout feature image, extract the feature vector using a convolutional neural network model, and then calculate the similarity between the feature vector in the visual element decorative design drawing and the feature vector of the input room layout image. The maximum similarity is selected as the visual element design scheme of the feature, and then the picture of the panoramic scheme is realized in order to let the designer get the customer’s feedback on the design effect in a more timely manner. The classification models developed in this paper achieve an accuracy rate exceeding 90%, effectively fulfilling the task of classifying images of room layout features. The user’s satisfaction with the interior design methodology experience projects proposed in this paper is also all higher than 70%, and the feedback from both designers and users is very good, which shows the excellent effect of deep convolutional networks in the optimization of interior design visual elements.

Thermal comfort performance prediction method using sports center layout images in several cold cities based on CNN

Sports centers are core spaces for urban sports and fitness, and reasonable layout of large public buildings can alleviate urban climate issues. But the adverse effects of planning and layout and surrounding urban microclimate on the thermal comfort of outdoor activity space are rarely considered. In this study, an EnergyPlus and Openfoam collaborative computing platform was constructed to calculate the universal thermal climate index of the outdoor space of sports center. By comparing the simulation results of sports centers in three severely cold cities, improving the wind environment of the site's microclimate was found to be key in improving thermal comfort, and the hours of overall site thermal comfort increased mainly in spring and autumn. In this paper, a design method combining numerical simulations and a convolutional neural network was proposed, which can effectively predict the thermal comfort performance of affected sites. The application of this prediction method to the design of the Yingkou Sports Center increased the hours of overall site thermal comfort by 15.01% compared with the original scheme. Synthesizing the results and methods of this study can provide an effective reference and analysis tool for the overall environmental performance of urban areas.

Research Progress on the Aesthetic Quality Assessment of Complex Layout Images Based on Deep Learning

With the development of the information age, the layout image is no longer a simple combination of text and graphics, but covers the complex layout image obtained from text, graphics, images and other layout elements through the process of artistic design, pre-press processing, typesetting, and so on. At present, the field of aesthetic-quality assessment mainly focuses on photographic images, and the aesthetic-quality assessment of complex layout images is rarely reported. However, the design of complex layout images such as posters, packaging labels, advertisements, etc., cannot be separated from the evaluation of aesthetic quality. In this paper, layout analysis is performed on complex layout images. Traditional and deep-learning-based methods for image layout analysis and aesthetic-quality assessment are reviewed and analyzed. Finally, the features, advantages and applications of common image aesthetic-quality assessment datasets and layout analysis datasets are compared and analyzed. Limitations and future perspectives of aesthetic assessment of complex layout images are discussed in relation to layout analysis and aesthetic characteristics.

Shooting Simulator «Inhibitor»: Target Layout Mathematical Support

Mathematical support of target layout with pre-scaling during storage and mixing of target image boundaries is described when displaying on a shooting range for the optical-electronic shooting simulator 'Inhibitor' developed at Institute of mechanics UdmFRC UB RAS and at Computer Engineering department of Kalashnikov ISTU jointly with JSC «Kalashnikov» Concern». The tactical and technical task for the implementation of the target layout is given: support for all targets from the Shooting Course and realistic targets, the angular dimensions of objects on the screen must correspond to the real ranges to them, and the speed of visualization algorithms allows simultaneously displaying at least 32 targets with a frequency of at least 25 frames/s. When developing mathematical support, the BMP format for storing TrueColor images with the scale of 1 pixel = 1cm2 was chosen, scaling targets to speed up further processing and the obligatory mixing of object-background border colors when displaying fractional coordinate support with pixel fractions of greatly reduced targets on the screen (so a thoracic target 50 cm wide at a distance of 100 m should be 12.8 points wide, while at a distance of 600 m - 2.13 points only) was emphasizes. The carried out studies of mathematical support performance for the target layout images formation confirmed compliance with the tactical and technical task requirements and recommended a pre-scaling factor of 0.5 (1:2) for optimal «quality-performance» ratio. Based on the prospects for further research and development of electronic shooting simulators and due to the improvement of computing tools and the development of software graphics libraries, it is necessary to expand the eficiency of simulators and reduce costs, which means to increase competitiveness, for example, by increasing the target layout display realism.

Study on the application of multi-angle imaging related technology in the construction process

Abstract The mufti-angle imaging technique is proposed to address the wall position offset and the side flaps and gate flaps caused by the sparse array in the ultra-wide band through-wall radar building layout imaging, using the characteristic that the azimuth upward distance profile is similar around the main flap but different around the side flap/gate flap. Firstly, after the backward projection algorithm obtains the wall image of each viewpoint, the mufti-directional two-dimensional matching filter is used to enhance the image of each viewpoint and eliminate the effect of side flaps and raster flaps. Then the images of each viewpoint are decomposed by combining the multistage transform, i.e., the non-down sampling contour let transform, and the obtained low-frequency components and high-frequency components are fused using the corresponding fusion rules, respectively. Finally, the final fused building layout image is obtained by non-down sampling contour let inverse transform. The simulation experimental data show that the running time of the method in this paper is 10.99s. It shows that the proposed method enhances the outline and detail information of the building wall, and the final fused image obtained has prominent wall image, clearer edge contours, and good retention of overall detail information. The quality of image fusion is greatly improved than other methods, which paves the way for subsequent fusion methods including chatterers such as construction building corners and windows and doors.

Transformer-based image generation from scene graphs

Graph-structured scene descriptions can be efficiently used in generative models to control the composition of the generated image. Previous approaches are based on the combination of graph convolutional networks and adversarial methods for layout prediction and image generation, respectively. In this work, we show how employing multi-head attention to encode the graph information, as well as using a transformer-based model in the latent space for image generation can improve the quality of the sampled data, without the need to employ adversarial models with the subsequent advantage in terms of training stability.The proposed approach, specifically, is entirely based on transformer architectures both for encoding scene graphs into intermediate object layouts and for decoding these layouts into images, passing through a lower dimensional space learned by a vector-quantized variational autoencoder. Our approach shows an improved image quality with respect to state-of-the-art methods as well as a higher degree of diversity among multiple generations from the same scene graph. We evaluate our approach on three public datasets: Visual Genome, COCO, and CLEVR. We achieve an Inception Score of 13.7 and 12.8, and an FID of 52.3 and 60.3, on COCO and Visual Genome, respectively. We perform ablation studies on our contributions to assess the impact of each component. Code is available at https://github.com/perceivelab/trf-sg2im.

End-to-end aperture layout optimization and image restoration for optical sparse aperture systems.

The optical sparse aperture (OSA) imaging technique is capable of improving the spatial resolution of a telescope while maintaining lower size, weight, and cost. The majority of OSA system researches separately focus on the design optimization of aperture layout and the method for image restoration, which have great design redundancy. In this Letter, an end-to-end design framework that simultaneously optimizes the aperture layout parameters of the OSA system and neural network parameters of image restoration is proposed, which achieves excellent imaging quality. The results show that adequate image mid-frequency information captured by the OSA system benefits network processing more than incomplete high-frequency information in a few directions. Based on this framework, we design a simplified OSA system on geostationary orbit. The simulation results show that our simplified OSA system with six sub-apertures measuring 1.2m each has a comparable imaging performance to a single-aperture system measuring 12 m.

Aortic Stenosis: Haemodynamic Benchmark and Metric Reliability Study

Aortic stenosis is a condition which is fatal if left untreated. Novel quantitative imaging techniques which better characterise transvalvular pressure drops are being developed but require refinement and validation. A customisable and cost-effective workbench valve phantom circuit capable of replicating valve mechanics and pathology was created. The reproducibility and relationship of differing haemodynamic metrics were assessed from ground truth pressure data alongside imaging compatibility. The phantom met the requirements to capture ground truth pressure data alongside ultrasound and magnetic resonance image compatibility. The reproducibility was successfully tested. The robustness of three different pressure drop metrics was assessed: whilst the peak and net pressure drops provide a robust assessment of the stenotic burden in our phantom, the peak-to-peak pressure drop is a metric that is confounded by non-valvular factors such as wave reflection. The peak-to-peak pressure drop is a metric that should be reconsidered in clinical practice.Graphical abstractThe left panel shows manufacture of low cost, functional valves. The central section demonstrates circuit layout, representative MRI and US images alongside gross valve morphologies. The right panel shows the different pressure drop metrics that were assessed for reproducibility

Trainable die-to-database for large field of view e-beam inspection

Extreme ultraviolet lithography has advanced microfabrication of semiconductor devices toward the sub-10-nm generation. In this situation, stochastic defects increase and hence process evaluation requires an entire wafer inspection at high speed. To satisfy this requirement, a large field of view (FoV) inspection with low-resolution enables us to inspect an entire wafer within an acceptable time because the throughput of e-beam inspection depends on imaging resolution. However, low-resolution images are difficult to inspect at high precision using conventional methods because of a smaller photographed defect size and worse signal-to-noise ratio. Moreover, deformation caused by the manufacturing process and larger distortion caused by large FoV result in false detections when we apply die-to-database (D2DB) inspection. To solve these issues, we propose trainable D2DB inspection, which predicts a pixel-value distribution of normal images from a corresponding design layout. The proposed method is robust to low-resolution images because it considers noise and acceptable deformation as variance of the learned distribution. In addition, by introducing a model to predict a misalignment between a design layout and inspection image, trainable D2DB becomes robust to image distortion. Experiments show that trainable D2DB can perform high-precision inspection on images with large noise and image distortion.

ПАРАМЕТРИЧЕСКИЕ БИБЛИОТЕЧНЫЕ ЭЛЕМЕНТЫ КАК ЭФФЕКТИВНОЕ СРЕДСТВО СОВЕРШЕНСТВОВАНИЯ ТЕХНОЛОГИЙ ИНФОРМАЦИОННОГО МОДЕЛИРОВАНИЯ В СТРОИТЕЛЬСТВЕ

It is rational to build automation of design activities as a stage of information modeling in construction on the formalization and modeling of numerous routine operations that make up the bulk of the complexity and cost of design. At the same time, elemental design is associated with frequent reference to specialized reference and regulatory literature, software external to the modeling environment, which form the necessary information flow for high-quality design. One of the effective directions for accelerating and improving design activities is the development and use of parametric models of elements of buildings, structures and building structures, containing both algorithms for constructing spatial and layout images of an object, and providing it with the necessary algorithmic equipment for automated design according to a normative or user-adapted scenario. The paper presents an example of substantiation, development and use of parametric library elements in information modeling of buildings, structures and building structures. On the example of a parametric model of a free-standing foundation, modeled in the Graphisoft GDL environment for use in the ARCHICAD PC, the goals of creating and the benefits of using parametric (smart) objects of the information model are considered, effective work with a parametric model is demonstrated, which reduces the complexity of design activities, the key parameters of the object and examples of scripts are given software environment, an assessment of efficiency was made, prospects for further development of parametric modeling tools were determined.

Traffic Flow Speed Prediction in Overhead Transport Systems for Semiconductor Fabrication Using Dense-UNet

To improve semiconductor productivity, efficient operation of the overhead hoist transport (OHT) system, which is an automatic wafer transfer device in a semiconductor fabrication plant (“fab”), is very important. A large amount of data is being generated in real time on the production line through the recent production plan of a smart factory. This data can be used to increase productivity, which in turn enables companies to increase their production efficiency. In this study, for the efficient operation of the OHT, the problem of OHT congestion prediction in the fab is addressed. In particular, the prediction of the OHT transport time was performed by training the deep convolutional neural network (CNN) using the layout image. The data obtained from the simulation of the fab and the actual logistics schedule data of a Korean semiconductor factory were used. The data obtained for each time unit included statistics on volume and speed. In the experiment, a layout image was created and used based on the statistics. The experiment was conducted using only the layout image without any other feature extraction, and it was shown that congestion prediction in the fab is effective.

Supervised Feature Extraction and Synthesis of Integrated Circuits Micrographs for Physical Assurance

Abstract This article proposes a design for a real-time Trojan detection system and explores possible solutions to the challenge of large-scale SEM image acquisition. One such solution, a deep-learning approach that generates synthetic micrographs from layout images, shows significant promise. Learning-based approaches are also used to both synthesize and classify cells. The classification outcome is matched with the design exchange format file entry to ensure the purity of the underlying IC.

Precise Pattern Alignment for Die-to-Database Inspection Based on the Generative Adversarial Network

Alignment between the reference layout (or target pattern) and the corresponding scanning electron microscope (SEM) image is a crucial task for the die-to-database (D2DB) inspection in the semiconductor industry. However, it is challenging to align them accurately because the style and quality of reference layouts represented as a computer-aided design (CAD) are quite different from those of grayscale SEM images with noise. Direct application of conventional cross-correlation-based matching methods often leads to misalignment. Here, we propose a new method enabling the precise pattern alignment. The main idea is to transform SEM images into target-like CAD images using a generative adversarial network (GAN). As the generated and real layout images have similar style and quality, they can be precisely aligned using conventional matching methods. Then, the SEM image can be located correctly on the corresponding reference layout for inspection. A polygon-based clustering algorithm for target patterns is developed to avoid manual selection and minimize the number of training data.

A Machine Learning Approach for Predicting Human Preference for Graph Layouts

Understanding what graph layout human prefer and why they prefer such graph layout is significant and challenging due to the highly complex visual perception and cognition system in the human brain. In this paper, we present the first machine learning approach for predicting human preference for graph layouts. Specifically, we propose a CNN-Siamese-based model to predict human preference from a pair of different layouts of the same graph. We employ a transfer learning method to overcome the insufficiency of the available ground truth human preference experiment data for training deep neural networks. Specifically, we exploit the quality metrics, which are correlated to human preference on graph layouts, to pre-train our model. Then, we fine-tune the model using the ground truth human preference experiment data. Experimental results using the ground truth human preference data sets show that our model M+HP can successfully predict human preference for graph layouts, achieving the average test accuracy of $92.28\%$ for large scale-free and mesh graphs. To our best knowledge, this is the first approach for predicting qualitative evaluation of graph layouts based on the ground truth human preference experiment data. Moreover, comparison experiments show that our model outperforms a simple baseline model and a previous Siamese-based model, demonstrating the importance of using graph layout images and the CNN-based model for predicting human preference.

Litho-NeuralODE 2.0: Improving hotspot detection accuracy with advanced data augmentation, DCT-based features, and neural ordinary differential equations

It has proved that the application of deep neural networks has advantage in lithographic hotspot detection, which is vital in the physical verification flow to reduce manufacturing yield loss. In this paper, we employ the discrete cosine transform (DCT)-based feature extraction method along with parameter search to compress the layout image to achieve higher classification accuracy and speed up the training process. To further improve the classification performance, data augmentation techniques addressing the imbalanced dataset problem along with neural ordinary differential equations based Litho-NeuralODE 2.0 framework with improved loss function have utilized in the work. Experimental results demonstrate that the proposed framework outperforms the state-of-the-art works with the lowest misses of 7 and the highest accuracy of 98.9% on average.

Automated Textile Circuit Generation using Machine Vision and Embroidery Technique

An automated textile circuit generation method was developed using machine vision and embroidery technique. For this, an image acquisition device was designed to capture the physical layout image of electronic devices on the fabric. A series of image analysis algorithms were developed to calibrate the image and recognize the type, location, and orientation of each device from the image. Dedicated computer-aided design software was developed that can design a circuit and modify it according to the actual device layout on the fabric. It can also convert the circuit elements such as conductive path and insulation layer into data for a computerized embroidery machine. An example circuit consisting of seven electronic devices was designed and repeatedly produced to verify the developed method. The method has shown promising results.

A retrieval and ranking method of mathematical documents based on CA-YOLOv5 and HFS.

In a retrieval system for mathematical documents based on mathematical expressions, the input and matching of mathematical expressions are key steps that affect the system's usability, accessibility and efficiency because of their special attributes. Therefore, this paper mainly focuses on improving the input efficiency and matching accuracy of mathematical expressions. This paper proposes a method for retrieval and ranking of mathematical documents based on CA-YOLOv5 and HFS (hesitation fuzzy set) by utilizing the advantages of CA (coordinate attention) model and YOLOv5 in target detection and the superiority of HFS in multiattribute decision-making. By embedding the CA model into the YOLOv5 network, the mathematical expressions in layout images are extracted and recognized to form mathematical query expressions. These expressions are then analyzed to obtain similarity evaluation features and matched with the candidate mathematical expressions indexed with the same features in a library of mathematical documents by employing the HFS as the similarity evaluation measure. Experiments were performed based on the TFD-ICDAR2019v2 dataset and the NTCIR dataset. The F1-score of the mathematical expression detection result was 76.54%, the MAP (mean average precision) of the mathematical documents retrieval result was 71.73%, and the average nDCG of mathematical documents ranking was 80.89%.

Construction of Multiobjective Planning Decision-Making Model of Ecological Building Spatial Layout under the Background of Rural Revitalization.

In order to improve the spatial layout ability of ecological buildings under the background of rural revitalization, a multiobjective planning and decision-making model for the spatial layout of ecological buildings is constructed. Based on the visual impact detection of ecological building space, a three-dimensional rendering model is established. The block matrix matching and boundary contour parameter analysis methods are used to plan and design the layout boundary feature points, and the wavelet scale decomposition method is used to analyze the mixed tone of the layout image. Based on this, a multiobjective planning decision-making model for the spatial layout of ecological buildings is established, based on which the spatial layout design scheme of ecological buildings is output to realize the spatial layout planning of ecological buildings. The simulation results show that the spatial layout of ecological buildings using this method is more reasonable, and the expression ability of ecological aesthetics is stronger, which has a good application value in rural planning and design.

Multimedia

Multimedia