Search

This paper describes three-dimensional (3D) reconstruction processes introducing an image selection method to efficiently generate a 3D model from an image sequence. To obtain suitable images for efficient 3D reconstruction, we applied the image selection method to remove redundant images in an image sequence. The proposed method can select suitable images from an image sequence based on optical flow measures and a fixed threshold. As a result, it can reduce the computational cost for 3D reconstruction processes based on the image sequence acquired by a camera. We confirmed that the computational cost of 3D reconstruction processes can be reduced while maintaining the 3D reconstruction accuracy at a constant level. [ABSTRACT FROM AUTHOR]

Remote sensing images often need to be merged into a larger mosaic image to support analysis on large areas in many applications. However, the performance of the mosaic imagery may be severely restricted if there are many areas with cloud coverage or if these images used for merging have a long-time span. Therefore, this paper proposes a method of image selection for full coverage image (i.e. a mosaic image with no cloud-contaminated pixels) generation. Specifically, a novel High-Frequency-Aware (HFA)-Net based on Swin-Transformer for region quality grading is presented to provide a data basis for image selection. Spatiotemporal constraints are presented to optimize the image selection. In the temporal dimension, the shortest-time-span constraint shortens the time span of the selected images, obviously improving the timeliness of the image selection results (i.e. with a shorter time span). In the spatial dimension, a spatial continuity constraint is proposed to select data with better quality and larger area, thus improving the radiometric continuity of the results. Experiments on the GF-1 images indicate that the proposed method reduces the averages by 76.1% and 38.7% in terms of the shortest time span compared to the Improved Coverage-oriented Retrieval algorithm (MICR) and Retrieval Method based on Grid Compensation (RMGC) methods, respectively. Moreover, the proposed method also reduces the residual cloud amount by an average of 91.2%, 89.8%, and 83.4% when compared to the MICR, RMGC, and Pixel-based Time-series Synthesis Method (PTSM) methods, respectively.

Remote sensing images often need to be merged into a larger mosaic image to support analysis on large areas in many applications. However, the performance of the mosaic imagery may be severely restricted if there are many areas with cloud coverage or if these images used for merging have a long-time span. Therefore, this paper proposes a method of image selection for full coverage image (i.e. a mosaic image with no cloud-contaminated pixels) generation. Specifically, a novel High-Frequency-Aware (HFA)-Net based on Swin-Transformer for region quality grading is presented to provide a data basis for image selection. Spatiotemporal constraints are presented to optimize the image selection. In the temporal dimension, the shortest-time-span constraint shortens the time span of the selected images, obviously improving the timeliness of the image selection results (i.e. with a shorter time span). In the spatial dimension, a spatial continuity constraint is proposed to select data with better quality and larger area, thus improving the radiometric continuity of the results. Experiments on the GF-1 images indicate that the proposed method reduces the averages by 76.1% and 38.7% in terms of the shortest time span compared to the Improved Coverage-oriented Retrieval algorithm (MICR) and Retrieval Method based on Grid Compensation (RMGC) methods, respectively. Moreover, the proposed method also reduces the residual cloud amount by an average of 91.2%, 89.8%, and 83.4% when compared to the MICR, RMGC, and Pixel-based Time-series Synthesis Method (PTSM) methods, respectively. [ABSTRACT FROM AUTHOR]

Karyotyping requires chromosome instances to be segmented and classified from the metaphase images. One of the difficulties in chromosome segmentation is that the chromosomes are randomly positioned in the image, and there is a great chance for chromosomes to be touched or overlap with others. It is always much easier for operators and automatic programs to tackle images without overlapping chromosomes than ones with largely overlapped chromosomes. In order to reduce the processing difficulty, adding a smart image selection procedure ahead of segmentation is practical and necessary. In this paper, we introduce the Smart Karyotyping Image Selection (SKIS) based on Commonsense Knowledge Reasoning. The initial experiment demonstrates that the proposed approach can select the expected images based on reasoning and benefit following karyotyping processes. [ABSTRACT FROM AUTHOR]

Asphalt mixtures exhibit complex mechanical behaviors due to their multiphase internal structures. To provide better characterizations of asphalt pavements under various forms of potential distress, a two-dimensional (2D) finite element simulation based on images of asphalt mixtures can be used to increase computational efficiency and reduce labor consumption. Nonetheless, using a representative image to eliminate the influence of dimension reduction from three dimensions to two dimensions is of great significance for attaining a reliable simulation result. Therefore, in this study, we investigated the consequence of dimension reduction for open-graded asphalt mixtures (denoted as OGFC-16), including a comprehensive characterization of these 2D models in terms of their morphologies and the similarities between them. This study aimed to reveal the variation in a 2D finite element simulation when applied to open-graded asphalt mixtures. Structural compositions, gradations, the aspect ratios of aggregates, and aggregate orientations were counted and calculated. In addition, the cosine similarity and structural similarity index measure (SSIM) were also calculated. Consequently, we performed a statistical analysis on the aforementioned indicators to quantitatively identify the discrepancy in the 2D images caused by dimension reduction. The results demonstrate that this 2D simulation might not be sufficient for representing the realistic mechanical performance of asphalt mixtures due to the remarkable variations in the image morphologies in different 2D images. However, the basic rules of stress behavior within structures can be accurately simulated. A compensative methodology for conducting a 2D simulation of open-graded asphalt mixtures should be based on a morphological characterization, considering structural compositions and the structural similarity index measure. [ABSTRACT FROM AUTHOR]

Vocal Cord Ulcer (VCU) is a contact ulcer that decreases the musculoskeletal laryngeal tension when speaking. With the advanced technology, including high-decision cameras and computational energy, it appears to be easy to construct. However, identifying laryngeal variation caused by VCU in CT images is still problematic. The paper aims to use image processing techniques to quantify the laryngeal variation caused by VCU, determine, and analyze its severity. The proposed 3D Swin Transforms Volumetric Segmentation Network (STVSNet) reduces the entanglement and improves the segmentation accuracy. Volumetric quantification on Contrast-Enhanced computed tomography (CECT) uses 3D STVSNet to extract shapes feature to evaluate the VCU severity. Evaluation results were 96.20% sensitivity, 97.15% accuracy, and 96.16% specificity. Concomitantly compared different prevail methods show better results for quantitative data. Experimental results show that 3D STVSNet indicates precise segmentation results for detecting VCU in any image type. [ABSTRACT FROM AUTHOR]

Task-oriented multimodal dialogue systems have important application value and development prospects. Existing methods have made significant progress, but the following challenges still exist: (1) Most existing methods focus on improving the accuracy of dialogue state tracking and dialogue act prediction. However, the essential to leverage knowledge in the knowledge base to supplement textual responses in multi-turn dialogues is ignored. (2) One feature that distinguishes multimodal dialogue from plain text dialogue is the usage of visual information. However, existing methods ignore the importance of accurately providing visual information to improve user satisfaction. (3) For multimodal dialogue systems, most existing methods ignore the classification of response types to assign appropriate response generators automatically. To address the issues above, we present a user-satisfactory multimodal dialogue system, USMD for short. Specifically, USMD is designed as four modules. The general response generator is based on generative pre-training 2.0 (GPT-2) to generate dialogue acts and general textual responses. The knowledge-enriched response generator is designed to leverage a structured knowledge base under the guidance of a knowledge graph. The image recommender pays attention to both latent and explicit visual cues, a deep multimodal fusion model to obtain informative image representations. Finally, the response classifier automatically selects the appropriate generators to answer the user based on user and agent actions. Extensive experiments on the benchmark multimodal dialogue datasets show that the proposed USMD model achieves state-of-the-art performance. [ABSTRACT FROM AUTHOR]

Asphalt mixtures exhibit complex mechanical behaviors due to their multiphase internal structures. To provide better characterizations of asphalt pavements under various forms of potential distress, a two-dimensional (2D) finite element simulation based on images of asphalt mixtures can be used to increase computational efficiency and reduce labor consumption. Nonetheless, using a representative image to eliminate the influence of dimension reduction from three dimensions to two dimensions is of great significance for attaining a reliable simulation result. Therefore, in this study, we investigated the consequence of dimension reduction for open-graded asphalt mixtures (denoted as OGFC-16), including a comprehensive characterization of these 2D models in terms of their morphologies and the similarities between them. This study aimed to reveal the variation in a 2D finite element simulation when applied to open-graded asphalt mixtures. Structural compositions, gradations, the aspect ratios of aggregates, and aggregate orientations were counted and calculated. In addition, the cosine similarity and structural similarity index measure (SSIM) were also calculated. Consequently, we performed a statistical analysis on the aforementioned indicators to quantitatively identify the discrepancy in the 2D images caused by dimension reduction. The results demonstrate that this 2D simulation might not be sufficient for representing the realistic mechanical performance of asphalt mixtures due to the remarkable variations in the image morphologies in different 2D images. However, the basic rules of stress behavior within structures can be accurately simulated. A compensative methodology for conducting a 2D simulation of open-graded asphalt mixtures should be based on a morphological characterization, considering structural compositions and the structural similarity index measure.

COVID-19 has spiked worldwide, having multiple outbreaks even with the production of vaccines. Imaging exams, such as Computer Tomography (CT) and X-ray, are recommended by the World Health Organization after performing RT-PCR tests to enhance COVID-19 diagnosis for serious cases. This work proposes a deep learning methodology to evaluate whether a patient presents COVID-19-related findings in CT images as an auxiliary diagnostic tool. As a CT exam produces many images related to a patient, some are irrelevant for COVID-19 diagnosis (e.g., closed lungs), using the raw information may hinder the model. Hence, we provide a CT scan image selection algorithm to filter the most informative images with two versions: (a) non-sequential, and (b) sequential. Then, online data augmentation is applied before feeding these images to a Convolutional Neural Network (CNN). Moreover, we evaluate the performance of the model for both versions of the CT selection algorithm in different approaches: (i) 'per-image', (ii) 'per-patient majority voting', and (iii) 'per-patient conservative voting (30%)'. We applied the proposed methodology in a Brazilian university hospital, a reference for COVID-19 treatment. For the test set, approaches (a-i), (a-ii), and (a-iii) reached an accuracy of 84.6%, 70%, and 70%, respectively, while approaches (b-i), (b-ii), and (b-iii) reached 90.9%, 80%, and 80%, respectively. Hence, we consider the proposed sequential version the most suitable image selection algorithm for the analyzed data set and useful for supporting decisions in COVID-19 diagnosis. [ABSTRACT FROM AUTHOR]

Although multimodal remote sensing data analysis can strongly improve the characterization of physical phenomena on Earth’s surface, nonidealities and estimation imperfections between records and investigation models can limit its actual information extraction ability. In this article, we aim at predicting the maximum information extraction that can be reached when analyzing a given data set. By means of an asymptotic information theory-based approach, we investigate the reliability and accuracy that can be achieved under optimal conditions for multimodal analysis as a function of data statistics and parameters that characterize the multimodal scenario to be addressed. Our approach leads to the definition of two indices that can be easily computed before the actual processing takes place. Moreover, we report in this article how they can be used for operational use in terms of image selection in order to maximize the robustness of the multimodal analysis, as well as to properly design data collection campaigns for understanding and quantifying physical phenomena. Experimental results show the consistency of our approach. [ABSTRACT FROM AUTHOR]

BackgroundEndoscopic ultrasound (EBUS) strain elastography can diagnose intrathoracic benign and malignant lymph nodes (LNs) by reflecting the relative stiffness of tissues. Due to strong subjectivity, it is difficult to give full play to the diagnostic efficiency of strain elastography. This study aims to use machine learning to automatically select high-quality and stable representative images from EBUS strain elastography videos.MethodsLNs with qualified strain elastography videos from June 2019 to November 2019 were enrolled in the training and validation sets randomly at a quantity ratio of 3:1 to train an automatic image selection model using machine learning algorithm. The strain elastography videos in December 2019 were used as the test set, from which three representative images were selected for each LN by the model. Meanwhile, three experts and three trainees selected one representative image severally for each LN on the test set. Qualitative grading score and four quantitative methods were used to evaluate images above to assess the performance of the automatic image selection model.ResultsA total of 415 LNs were included in the training and validation sets and 91 LNs in the test set. Result of the qualitative grading score showed that there was no statistical difference between the three images selected by the machine learning model. Coefficient of variation (CV) values of the four quantitative methods in the machine learning group were all lower than the corresponding CV values in the expert and trainee groups, which demonstrated great stability of the machine learning model. Diagnostic performance analysis on the four quantitative methods showed that the diagnostic accuracies were range from 70.33% to 73.63% in the trainee group, 78.02% to 83.52% in the machine learning group, and 80.22% to 82.42% in the expert group. Moreover, there were no statistical differences in corresponding mean values of the four quantitative methods between the machine learning and expert groups (p >0.05).ConclusionThe automatic image selection model established in this study can help select stable and high-quality representative images from EBUS strain elastography videos, which has great potential in the diagnosis of intrathoracic LNs.

Background: Endoscopic ultrasound (EBUS) strain elastography can diagnose intrathoracic benign and malignant lymph nodes (LNs) by reflecting the relative stiffness of tissues. Due to strong subjectivity, it is difficult to give full play to the diagnostic efficiency of strain elastography. This study aims to use machine learning to automatically select high-quality and stable representative images from EBUS strain elastography videos. Methods: LNs with qualified strain elastography videos from June 2019 to November 2019 were enrolled in the training and validation sets randomly at a quantity ratio of 3:1 to train an automatic image selection model using machine learning algorithm. The strain elastography videos in December 2019 were used as the test set, from which three representative images were selected for each LN by the model. Meanwhile, three experts and three trainees selected one representative image severally for each LN on the test set. Qualitative grading score and four quantitative methods were used to evaluate images above to assess the performance of the automatic image selection model. Results: A total of 415 LNs were included in the training and validation sets and 91 LNs in the test set. Result of the qualitative grading score showed that there was no statistical difference between the three images selected by the machine learning model. Coefficient of variation (CV) values of the four quantitative methods in the machine learning group were all lower than the corresponding CV values in the expert and trainee groups, which demonstrated great stability of the machine learning model. Diagnostic performance analysis on the four quantitative methods showed that the diagnostic accuracies were range from 70.33% to 73.63% in the trainee group, 78.02% to 83.52% in the machine learning group, and 80.22% to 82.42% in the expert group. Moreover, there were no statistical differences in corresponding mean values of the four quantitative methods between the machine learning and expert groups (p >0.05). Conclusion: The automatic image selection model established in this study can help select stable and high-quality representative images from EBUS strain elastography videos, which has great potential in the diagnosis of intrathoracic LNs. [ABSTRACT FROM AUTHOR]

Copyright of Science-Based Technologies is the property of National Aviation University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Several images are used as a part of the interactive data in the Nipo-Brazilian digital TV system that require good subjective quality while using the lowest possible bandwidth but the well-known traditional compression systems are already applied by default to these images. The concept of a mosaic image (several images forming one) has been formerly used in the steganography application field. The mosaic image is obtained by reordering the image-blocks of a secret image disguised as another image, the so-called target image, and then feeding both to a near reversible color-transformation algorithm. Its use is a possible solution to this need in digital interactive television, for which the main challenge is to achieve this by using less bandwidth for transmission. We propose here a procedure to reduce the amount of data needed to recover the secret image from the mosaic image, as well as a criterion to select the target image and therefore improve the quality of the recovered secret image in interactive data applications. The main objective is to efficiently transmit two images as one using a lower bandwidth. On the one hand, the number of bits needed to recover a given secret image is highly reduced by modeling the image-block standard deviation statistical distribution. On the other hand, the entropy of the image-block means and standard deviations per color component are used to identify the most convenient target image among the images set present in the interactive application of interest. A series of experiments with a set of 20 mostly natural images showed a reduction in the number of bits close to 3-to-1 with respect to techniques of reference. The proposed method allows us to improve the bandwidth use by reducing the number of bits needed to recover the secret image, it preserves the subjective quality of recovered secret images, and it gives the possibility to determine the best target images available in several multimedia and digital terrestrial television applications.

With the prevalence of smartphones, the demand of recognizing the location through their camera and sensors is paid abundant attentions. For constructing a location recognition image database, the crowdsourcing technology is introduced to collect images associated with other sensory data. However, as abundant crowdsourced images evolve, it is essential to select high-quality images to decrease the burden of storage when designing an offline location recognition system directly on mobile devices. To address this problem, we propose an image selection framework, i.e., Informative image Selection Framework (ISF), considering both the diversity in spatial distribution and representativeness of images with high quality. First, for the images corresponding to the same object, we propose the Self-adaptive Space Clustering algorithm to group them into several clusters for maintaining high diversity of the image database. Second, for every cluster, we propose the Crucial Part Feature Detection algorithm to detect representative images. Extensive experiments demonstrate that ISF is effective and efficient for image selection, outperforming other similar image selection schemes around 5%. [ABSTRACT FROM AUTHOR]

The primary argument of this paper arises from the need identified in fashion design education within a South African context for a shift in the focus of theoretical curricula content to convert from a diploma to a degree structure. This paper addresses this concern and reports on the baseline assessment of a devised and implemented tool to advance visual literacy thinking skills of undergraduate fashion design students, applied to the selection of images for mood boards. The inquiry adopted an action research design and obtained data from systematically documented reflective notes, completed student image analysis help sheets and a comparison of assessment results of mood boards created before and after the application of the tool. While assessment results indicate an improvement in the level of thinking skills of some students, surface learning was identified as an underlying challenge to address for refinement of the tool. Improved results, nevertheless, suggest that the strategic principles facilitated through the tool and the subsequent recommendations for refinement present a potential framework to underpin the visual conceptual design challenges of students for consideration in a degree structure. [ABSTRACT FROM PUBLISHER]

In recent years, acquiring information of remote sensing data, especially satellite data has excessively increased and several methods are presented in order to improve the quality of remote sensing studies in earth sciences. It is possible to manage many projects and provide different types of thematic maps in a short period of time, and a low cost by utilizing satellite data and GIS method. Recent researches show that utilizing satellite data in studying natural phenomena can effectively help to reduce the time and cost at the same time maximize the precision. But, many users of these data face confusion at choosing suitable image for their subject and lack a special criterion for that end, Or else they merely take one or two criteria in to account and lack a comprehensive view in choosing the best image. Therefore, defining and analyzing criteria for correct and precise selection of satellite data, in accordance with case-study, is crucial. So, in this article, we investigate the image selection criteria, especially their role in minimizing time, cost and extracting useful data. On the basis of the results, prior to doing of the project, users of these data need to study selection criteria properly. After that, on the basis of these criteria and phenomena under study they should set out to choose sensor type, date of image acquisition, image type, and methods of information extraction. Therefore in research, different practical aspects of satellite images as well as criteria for selecting suitable images are investigated and subsequently information and suitable solutions are provided for users.

Images play an important role in the representation and acquisition of specialized knowledge. Not surprisingly, terminological knowledge bases (TKBs) often include images as a way to enhance the information in concept entries. However, the selection of these images should not be random, but rather based on specific guidelines that take into account the type and nature of the concept being described. This paper presents a proposal on how to combine the features of images with the conceptual propositions in EcoLexicon, a multilingual TKB on the environment. This proposal is based on the following: (1) the combinatory possibilities of concept types; (2) image types, such as photographs, drawings and flow charts; (3) morphological features or visual knowledge patterns (VKPs), such as labels, colours, arrows, and their effect on the functional nature of each image type. Currently, images are stored in association with concept entries according to the semantic content of their definitions, but they are not described or annotated according to the parameters that guided their selection, which would undoubtedly contribute to the systematization and automatization of the process. First, the images included in EcoLexicon were analyzed in terms of their adequateness, the semantic relations expressed, the concept types and their VKPs. Then, with these data, guidelines for image selection and annotation were created. The final aim is twofold: (1) to systematize the selection of images and (2) to start annotating old and new images so that the system can automatically allocate them in different concept entries based on shared conceptual propositions. [ABSTRACT FROM AUTHOR]

3D models have been widely used by spread of many available free-software. Additionally, enormous images can be easily acquired, and images are utilized for creating the 3D models recently. The creation of 3D models by using huge amount of images, however, takes a lot of time and effort, and then efficiency for 3D measurement are required. In the efficient strategy, the accuracy of the measurement is also required. This paper develops an image selection method based on network design that means surveying network construction. The proposed method uses image connectivity graph. The image connectivity graph consists of nodes and edges. The nodes correspond to images to be used. The edges connected between nodes represent image relationships with costs as accuracies of orientation elements. For the efficiency, the image connectivity graph should be constructed with smaller number of edges. Once the image connectivity graph is built, the image selection problem is regarded as combinatorial optimization problem and the graph cuts technique can be applied. In the process of 3D reconstruction, low quality images and similar images are also extracted and removed. Through the experiments, the significance of the proposed method is confirmed. It implies potential to efficient and accurate 3D measurement. [ABSTRACT FROM AUTHOR]

Benthic imagery is an effective tool for quantitative description of ecologically and economically important benthic habitats and biota. The recent development of autonomous underwater vehicles (AUVs) allows surveying of spatial scales that were previously unfeasible. However, an AUV collects a large number of images, the scoring of which is time and labour intensive. There is a need to optimise the way that subsamples of imagery are chosen and scored to gain meaningful inferences for ecological monitoring studies. We examine the trade-off between the number of images selected within transects and the number of random points scored within images on the percent cover of target biota, the typical output of such monitoring programs. We also investigate the efficacy of various image selection approaches, such as systematic or random, on the bias and precision of cover estimates. We use simulated biotas that have varying size, abundance and distributional patterns. We find that a relatively small sampling effort is required to minimise bias. An increased precision for groups that are likely to be the focus of monitoring programs is best gained through increasing the number of images sampled rather than the number of points scored within images. For rare species, sampling using point count approaches is unlikely to provide sufficient precision, and alternative sampling approaches may need to be employed. The approach by which images are selected (simple random sampling, regularly spaced etc.) had no discernible effect on mean and variance estimates, regardless of the distributional pattern of biota. Field validation of our findings is provided through Monte Carlo resampling analysis of a previously scored benthic survey from temperate waters. We show that point count sampling approaches are capable of providing relatively precise cover estimates for candidate groups that are not overly rare. The amount of sampling required, in terms of both the number of images and number of points, varies with the abundance, size and distributional pattern of target biota. Therefore, we advocate either the incorporation of prior knowledge or the use of baseline surveys to establish key properties of intended target biota in the initial stages of monitoring programs. [ABSTRACT FROM AUTHOR]