Search

The accurate detection and counting of flowers ensure the grading quality of the ornamental plants. In automated potted flower grading scenarios, low detection precision, occlusions and overlaps impact counting accuracy. This study proposed a counting method combining a deep learning algorithm with multiple viewpoints. Firstly, a flower detection model, PA-YOLO, was developed based on YOLOv5 by designing a two-scale detection branch, optimizing the number of bottlenecks and integrating a dynamic head framework. Next, PA-YOLO was used to detect grouped 360-viewpoint images of each potted plant to determine the optimal number of viewpoints for counting. The detection results indicated that PA-YOLO achieved a mean average precision (mAP) of 95.4% and an average precision (AP) of 91.9% for occluded blooms on our Phalaenopsis flower dataset. For the optimal number of viewpoints, the average counting accuracy of buds and blooms was highest at three viewpoints, with scores of 96.25% and 93.33%, respectively. The final counting accuracy reached 95.56% in flower counting tests conducted from three viewpoints. The overall results suggest that the proposed method can effectively detect and count flowers in complex occlusion and overlap environments, providing guidance for designing and implementing the vision component in an automated potted flower grading system. [ABSTRACT FROM AUTHOR]

Data clustering is an unsupervised technique that can be used to partition the data into groups based on the similarities of the retrieved objects using different distance metrics like Euclidean, cosine, etc. In contrast to Euclidean, the cosine computes the objects similarity by considering both the magnitude and direction of the data vectors. As a result, it performed far better than a standard Euclidean distance metric in applications involving real-time data clustering. The initial k-value (clustering tendency) is required by top clustering techniques like k-means and hierarchical approaches to determine the clusters' quality. Users with knowledge can assign the k-value. However, sometimes the right k-value in such algorithms may need to be assigned. After a thorough review of the work, it was discovered that the visual technique known as visual assessment of (cluster) tendency (VAT) effectively addresses the clustering tendency issue. It uses the Euclidean metric to find the similarity features in its algorithm. Another enhanced visual technique, cosine-based VAT(cVAT), outperformed the VAT for text data and speech clustering applications. However, the similarity features are extracted about a single viewpoint in cVAT. This paper develops the multi-viewpoints-based cosine similarity measure (MVPCSM) for a more informative assessment. Instead of using a single reference point like a typical cosine measure, the MVPCSM generates precise similarity characteristics using several views. The performance of the existing and proposed technique (MVPCSM-VAT) is evaluated using clustering accuracy (CA) and normalized mutual information (NMI). It has been demonstrated that the proposed MVPCSM-VAT is 15-25% more efficient than VAT and cVAT in terms of the parameters of CA and NMI. The proposed method successfully obtains more quality data clusters than MVS-VAT. [ABSTRACT FROM AUTHOR]

Skeleton-based gesture recognition (SHGR) is a very challenging task due to the complex articulated topology of hands. Previous works often learn hand characteristics from a single observation viewpoint. However, the various context information hidden in multiple viewpoints is disregarded. To resolve this issue, we propose a novel multi-view hierarchical aggregation network for SHGR. Firstly, two-dimensional non-uniform spatial sampling, a novel strategy forming extrinsic parameter distributions of virtual cameras, is presented to enumerate viewpoints to observe hand skeletons. Afterward, we adopt coordinate transformation to generate multi-view hand skeletons and employ a multi-branch convolutional neural networks to further extract the multi-view features. Furthermore, we exploit a novel hierarchical aggregation network including hierarchical attention architecture and global context modeling to fuse the multi-view features for final classification. Experiments on three benchmarked datasets demonstrate that our work can be competitive with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The accurate detection and counting of flowers ensure the grading quality of the ornamental plants. In automated potted flower grading scenarios, low detection precision, occlusions and overlaps impact counting accuracy. This study proposed a counting method combining a deep learning algorithm with multiple viewpoints. Firstly, a flower detection model, PA-YOLO, was developed based on YOLOv5 by designing a two-scale detection branch, optimizing the number of bottlenecks and integrating a dynamic head framework. Next, PA-YOLO was used to detect grouped 360-viewpoint images of each potted plant to determine the optimal number of viewpoints for counting. The detection results indicated that PA-YOLO achieved a mean average precision (mAP) of 95.4% and an average precision (AP) of 91.9% for occluded blooms on our Phalaenopsis flower dataset. For the optimal number of viewpoints, the average counting accuracy of buds and blooms was highest at three viewpoints, with scores of 96.25% and 93.33%, respectively. The final counting accuracy reached 95.56% in flower counting tests conducted from three viewpoints. The overall results suggest that the proposed method can effectively detect and count flowers in complex occlusion and overlap environments, providing guidance for designing and implementing the vision component in an automated potted flower grading system.

A method that combines Lucy-Richardson deconvolution algorithm (LRA) with a multiview projection setup to improve imaging reconstruction of objects embedded in scattering media is presented. In the first step, the medium was illuminated with a laser beam and single-shot multiple images of the object were obtained from different viewpoints. Next, the above procedure was performed under the same conditions but only the medium, without the object, was illuminated by a light point source to obtain the system PSF. Third, each sub-image of the object was digitally cropped, extracted from the array, and deconvolved with the average PSF image following the LRA. Finally, all the sub-deconvolved images were shifted to a common point and superimposed to reveal the shape of the hidden object. Various setup conditions including medium turbidity, thickness, target shape, and illumination state were tested and compared. Results indicate an effective performance of the proposed method. [ABSTRACT FROM AUTHOR]

Copyright of Chinese Journal of Liquid Crystal & Displays is the property of Chinese Journal of Liquid Crystal & Displays 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.)

Skeleton based human action recognition has attracted more and more attentions recently thanks to the accessibility of depth sensors and the development of pose estimation techniques. Conventional approaches such as convolutional neural networks usually model skeletons with grid-shaped representations, which cannot explicitly explore the dependency between two correlated joints. In this paper, we treat the skeleton as a single graph with joints as nodes and bones as edges. Based on the skeleton graph, the improved graph convolutional network called adaptive multi-view graph convolutional networks(AMV-GCNs) is proposed to deal with skeleton based action recognition. We firstly construct a novel skeleton graph and two kinds of graph nodes are defined to model the spatial configuration and temporal dynamics respectively. Then the generated graphs along with feature vectors on graph nodes are fed into AMV-GCNs. In AMV-GCNs, an adaptive view transformation module is designed to reduce the impact of view diversity. Proposed module can automatically determine suitable viewpoints and transform skeletons to new representations under those viewpoints for better recognition. Further, we employ multiple GCNs based streams to utilize and learn action information from different viewpoints. Finally, the classification scores from multiple streams are fused to provide the recognition result. Extensive experimental evaluations on four challenging datasets, NTU RGB+D 60, NTU RGB+D 120, Northwestern-UCLA and UTD-MHAD, demonstrate the superiority of our proposed network. [ABSTRACT FROM AUTHOR]

In this article, we view business models as complex deal-making activity systems organized to create, deliver, and capture value. Unlike some other viewpoints, we emphasize both system components and their interconnection. Business activities are carried out by a network of actors drawing on a network of resources, and individual firms seek to configure these intersecting networks to enhance their competitive positioning. The business model literature refers to the significance of antecedent activities in providing context – opportunities the firm decides to pursue, the strategy adopted, and requisite capabilities. Drawing on this literature, we propose an approach to framing business model context. Drawing on the information systems literature, we identify a toolkit facilitating activity system architecture design. We suggest how this both draws out the underlying complexity of a business model and shows how a multiplicity of views makes sense.

Background. Cancer survivors must manage a range of adverse symptoms and side effects postdiagnosis. These effects often co-occur with preexisting comorbid conditions. Recognizing the complex chronicity of the disease, self-management support interventions have been developed to promote cancer survivors' knowledge, skills, and confidence to self-manage their health. Though shown to be beneficial, the processes underlying self-management have yet to be explicated. Aim. To explore how a community-based self-management support intervention fosters cancer survivors' knowledge, confidence, and skills to self-manage their health. Method. A qualitative case study adopting multiple viewpoints was utilized. Seventeen cancer survivors who participated in a self-management support intervention (referred to as a cancer coaching program) were interviewed and six program staff took part in a focus group. Transcripts from the interviews and focus group were analyzed using a hybrid inductive–deductive approach guided by principles of qualitative description. Results. Cancer survivors and program staff offered complementary perspectives that enhanced our understanding of how the self-management support intervention fosters cancer survivors' knowledge, confidence, and skills to self-manage their health. Four themes captured strategies necessary to promote self-management: (1) looking beyond the disease: the importance of holistic person-centered care, (2) cocreating plans: the key to effective and meaningful self-management support, (3) fostering activation via tailored, targeted, and expert-sourced information and resources, and (4) having dependable and impartial emotional support. Within each theme, pertinent self-management support strategies were described by cancer survivors and program staff. Discussion and Conclusion. Findings offer an understanding of how one community-based self-management support intervention promotes self-management and highlight valuable self-management support strategies that could be incorporated into existing and future interventions. [ABSTRACT FROM AUTHOR]

Professor Duzheng YE's name has been familiar to me ever since my postdoctoral years at MIT with Professors Jule CHARNEY and Norman PHILLIPS, back in the late 1960s. I had the enormous pleasure of meeting Professor YE personally in 1992 in Beijing. His concern to promote the very best science and to use it well, and his thinking on multi-level orderly human activities, reminds me not only of the communication skills we need as scientists but also of the multi-level nature of science itself. Here I want to say something (a) about what science is; (b) about why multi-level thinking-and taking more than one viewpoint-is so important for scientific as well as for other forms of understanding; and (c) about what is meant, at a deep level, by 'scientific understanding' and trying to communicate it, not only with lay persons but also across professional disciplines. I hope that Professor YE would approve. [ABSTRACT FROM AUTHOR]

Copyright of Journal of International Social Research is the property of Journal of International Social Research 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.)

Alternative approaches to three-dimensional (3-D) reconstruction by employing the concepts of “system identification” and “communication systems” based on structured light patterns are proposed. In addition, a sampling criterion of the light source is derived in the case of using multiple projectors because 3-D reconstruction sometimes employs multiple viewpoints (cameras) and multiple structured light sources (or projectors). To reformulate a reconstruction problem, an input–output (I/O) system theoretic is adopted, and camera (s) and light source(s) that are located at different positions are defined as the output and the input, respectively. Akin to the system identification problem, the ratio of an output to an input, the “system function,” is defined as a 3-D measurement result. Alternatively, the reconstruction work can employ the concept of the “modulation and demodulation theory,” and the reconstruction work can be reinterpreted as an “input estimation problem.” This contribution chiefly deals with approximate reconstruction results that are sufficient for practical applications, such as 3-D object detection, clarification, recognition, and classification, rather than a perfect 3-D reconstruction itself. To that end, the development of an efficient and fast 3-D imaging system framework is proposed. [ABSTRACT FROM AUTHOR]

The goal of this paper is to identify individuals by analyzing their gait. Instead of using binary silhouettes as input data (as done in many previous works) we propose and evaluate the use of motion descriptors based on densely sampled short-term trajectories. We take advantage of state-of-the-art people detectors to define custom spatial configurations of the descriptors around the target person, obtaining a rich representation of the gait motion. The local motion features (described by the Divergence-Curl-Shear descriptor [M. Jain, H. Jegou and P. Bouthemy, Better exploiting motion for better action recognition, in Proc. IEEE Conf. Computer Vision Pattern Recognition (CVPR) (2013), pp. 2555-2562.]) extracted on the different spatial areas of the person are combined into a single high-level gait descriptor by using the Fisher Vector encoding [F. Perronnin, J. Sánchez and T. Mensink, Improving the Fisher kernel for large-scale image classification, in Proc. European Conf. Computer Vision (ECCV) (2010), pp. 143-156]. The proposed approach, coined Pyramidal Fisher Motion, is experimentally validated on 'CASIA' dataset [S. Yu, D. Tan and T. Tan, A framework for evaluating the effect of view angle, clothing and carrying condition on gait recognition, in Proc. Int. Conf. Pattern Recognition, Vol. 4 (2006), pp. 441-444]. (parts B and C), 'TUM GAID' dataset, [M. Hofmann, J. Geiger, S. Bachmann, B. Schuller and G. Rigoll, The TUM Gait from Audio, Image and Depth (GAID) database: Multimodal recognition of subjects and traits, J. Vis. Commun. Image Represent. 25(1) (2014) 195-206]. 'CMU MoBo' dataset [R. Gross and J. Shi, The CMU Motion of Body (MoBo) database, Technical Report CMU-RI-TR-01-18, Robotics Institute (2001)]. and the recent 'AVA Multiview Gait' dataset [D. López-Fernández, F. Madrid-Cuevas, A. Carmona-Poyato, M. Marín-Jiménez and R. Muñoz-Salinas, The AVA multi-view dataset for gait recognition, in Activity Monitoring by Multiple Distributed Sensing, Lecture Notes in Computer Science (Springer, 2014), pp. 26-39]. The results show that this new approach achieves state-of-the-art results in the problem of gait recognition, allowing to recognize walking people from diverse viewpoints on single and multiple camera setups, wearing different clothes, carrying bags, walking at diverse speeds and not limited to straight walking paths. [ABSTRACT FROM AUTHOR]

We experimentally demonstrate an integrated approach for improving reconstruction quality of targets embedded within a turbid media by integrating multiple viewpoints (projections) together with an extended unsharp masking (e USM) algorithm recently demonstrated in atmospheric solar images. In the setup, objects with differing geometrical shapes and sizes were embedded within media with a range of turbidity and varying attenuation and viscosity values. The medium was illuminated with a laser beam after which multiview images, obtained by lens array, were captured with a camera. In offline processing, each sub-image (viewpoint) was digitally cropped, extracted from the array, e USM filtered, and qualitatively evaluated by contrast metrics. After sub-images were sorted and ranked based on metric values, the selected 'best' sub-images were shifted to a common center and co-added with others using a shift-and-add (SAA) method to form a single average image revealing the shape of the object. Linear polarization was also introduced into the setup to give higher contrast image quality. Moreover, by optimizing e USM parameters, further improvement was achieved in the reconstruction. Experimental results illustrate the improvement of the proposed approach as compared to the previously described SAA method alone. [ABSTRACT FROM AUTHOR]

We study the problem of visibility in polyhedral terrains in the presence of multiple viewpoints. We consider a triangulated terrain with m > 1 viewpoints (or guards) located on the terrain surface. A point on the terrain is considered visible if it has an unobstructed line of sight to at least one viewpoint. We study several natural and fundamental visibility structures: (1) the visibility map, which is a partition of the terrain into visible and invisible regions; (2) the colored visibility map, which is a partition of the terrain into regions whose points have exactly the same visible viewpoints; and (3) the Voronoi visibility map, which is a partition of the terrain into regions whose points have the same closest visible viewpoint. We study the complexity of each structure for both 1.5D and 2.5D terrains, and provide efficient algorithms to construct them. Our algorithm for the visibility map in 2.5D terrains improves on the only existing algorithm in this setting. To the best of our knowledge, the other structures have not been studied before. [ABSTRACT FROM AUTHOR]

Jazz music is a genre that consists mainly of improvising over known tunes, represented as alead sheet. This study addresses the question ‘to what extent does a lead sheet carry information about its composer?’ Primarily, this study considers chord progressions alone, and secondarily melodic and temporal information combined with various multiple viewpoint models. Using these classifiers, a novel subsequence selection algorithm is presented to trace stylistic similarities within a lead sheet. We conclude that composers can, to a reasonable extent, be recognized from their chord progressions, and that the consideration of melodic and temporal information improves classification accuracy by a small but statistically significant amount. [ABSTRACT FROM AUTHOR]

We discuss the problem of automatic four-part harmonization: given a soprano part, add alto, tenor and bass in accordance with the compositional practices of a particular musical era. In particular, we focus on the development of representational and modelling techniques, within the framework ofmultiple viewpoint systemsandPrediction by Partial Match(PPM), for the creation of statistical models of four-part harmony by machine learning. Our ultimate goal is to create better models, according to the information theoretic measurecross-entropy, than have yet been produced. We use multiple viewpoint because of their ability to represent both surface and underlying musical structure, and because they have already been successfully applied to melodic modelling. To allow for the complexities of harmony, however, the framework must be extended; for example, we begin by predicting complete chords, and then extend the framework to allow part by part prediction. As the framework is extended and generalized, the viewpoints become more complex. This article discusses matters related to viewpointdomains(alphabets), such as their size and consequent effect on run time; and presents methods for their reliable construction. We also present an empirical analysis of the time complexity of our computer implementation. [ABSTRACT FROM AUTHOR]

Creativity is often needed in requirements elicitation, i.e., requirement idea generation; and techniques to enhance creativity are believed to be useful. This paper describes two controlled experiments to compare the requirements-elicitation effectiveness of three creativity enhancement techniques (CET): (1) full EPMcreate; (2) Power-Only EPMcreate, an optimization of full EPMcreate; and (3) traditional brainstorming. In each experiment, one team of university students applied one of the two or three CETs under study in the experiment to generate ideas for requirements for enhancing a high school's public Web site. The results of the first experiment indicate that Power-Only EPMcreate is more effective, by the quantity and quality of the ideas generated, than the full EPMcreate, which is, in turn, more effective than brainstorming. The results of the second experiment confirm that Power-Only EPMcreate is more effective, by the same measures, than full EPMcreate. In each experiment, for the sake of uniform, reproducible evaluation, a requirement idea is considered high quality if it is both new and useful. [ABSTRACT FROM AUTHOR]

We present micro cameras that provide multiple viewpoint imaging from one single image capture and without moving the camera. The micro cameras are based on lightfield photography where the camera captures the 4-D lightfield and images it onto a 2-D charge-coupled device (CCD) sensor through the use of a microlens array. The microlens array was fabricated using the photoresist reflow process. Two micro cameras were realized. One camera has a 15.2-mm aperture, a 97 \times 97 microlens array of 230- \mu\m pitch size, and a 1-megapixel CCD and is capable of 100 different viewpoints with a viewing angle of approximately 20^\circ. The other camera has a 9-mm aperture, a 56 \times 42 microlens array of 80-\mu\m pitch size, and a 5-megapixel CCD and is capable of around 1000 different viewpoints with a viewing angle of approximately 4.7 ^\circ. The resolution of the images rendered from the two cameras is 97 \times 97 and 56 \times 42 pixels, respectively.\hfill [2011-0314] [ABSTRACT FROM AUTHOR]

In many scenarios a dynamic scene is filmed by multiple video cameras located at different viewing positions. Visualizing such multi-view data on a single display raises an immediate question-which cameras capture better views of the scene? Typically, (e.g. in TV broadcasts) a human producer manually selects the best view. In this paper we wish to automate this process by evaluating the quality of a view, captured by every single camera. We regard human actions as three-dimensional shapes induced by their silhouettes in the space-time volume. The quality of a view is then evaluated based on features of the space-time shape, which correspond with limb visibility. Resting on these features, two view quality approaches are proposed. One is generic while the other can be trained to fit any preferred action recognition method. Our experiments show that the proposed view selection provide intuitive results which match common conventions. We further show that it improves action recognition results. [ABSTRACT FROM AUTHOR]

Viewpoint calibration is a method to manipulate hand-eye for generating calibration parameters for active viewpoint control and object grasping. In robot vision applications, accurate vision sensor calibration and robust vision-based robot control are essential for developing an intelligent and autonomous robotic system. This paper presents a new approach to hand-eye robotic calibration for vision-based object modeling and grasping. Our method provides a 1.0-pixel level of image registration accuracy when a standard Puma/Kawasaki robot generates an arbitrary viewpoint. To attain this accuracy, our new formalism of hand-eye calibration deals with a lens distortion model of a vision sensor. Our most distinguished approach of optimizing intrinsic parameters is to utilize a new parameter estimation algorithm using an extended Kalman filter. Most previous approaches did not even consider the optimal estimates of the intrinsic and extrinsic camera parameters, or chose one of the estimates obtained from multiple solutions, which caused a large amount of estimation error in hand-eye calibration. We demonstrate the power of this new method for: 1) generating 3-D object models using an interactive 3-D modeling editor; 2) recognizing 3-D objects using stereovision systems;, and 3) grasping 3-D objects using a manipulator. Experimental results using Puma and Kawasaki robots are shown. [ABSTRACT FROM AUTHOR]

The calculation of viewsheds is a routine operation in geographic information systems and is used in a wide range of applications. Many of these involve the siting of features, such as radio masts, which are part of a network and yet the selection of sites is normally done separately for each feature. The selection of a series of locations which collectively maximise the visual coverage of an area is a combinatorial problem and as such cannot be directly solved except for trivial cases. In this paper, two strategies for tackling this problem are explored. The first is to restrict the search to key topographic points in the landscape such as peaks, pits and passes. The second is to use heuristics which have been applied to other maximal coverage spatial problems such as location–allocation. The results show that the use of these two strategies results in a reduction of the computing time necessary by two orders of magnitude, but at the cost of a loss of 10% in the area viewed. Three different heuristics were used, of which Simulated Annealing produced the best results. However the improvement over a much simpler fast-descent swap heuristic was very slight, but at the cost of greatly increased running times. [Copyright &y& Elsevier]

In this paper, the authors use gesture images obtained from multiple cameras to estimate hand shapes. First, for the entered multiple viewpoint images, they use the outline in each separate image to extract features, create feature vectors, and calculate the difference between the image and each model image in the database. Next, they calculate the shape complexity as a value indicating the degree to which the image of each viewpoint is distinctive and use these values to integrate the individual differences to obtain the overall difference. They select the model image having the minimum integrated difference and let it be the estimation result. The authors also use a transition network to reduce the amount of searching during hand shape estimation. During the training stage, the authors used training image sequences to automatically register model images and possible transitions between model images to create a transition network. During the estimation stage, they used the transition network to perform matching more efficiently by limiting the candidates to only model images that can transition from the corresponding model image in the preceding frame. The authors performed experiments using binocular images to estimate hand shapes and showed the results of matching using the transition network. © 2004 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 87(8): 33–42, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjc.20089 [ABSTRACT FROM AUTHOR]

This paper proposes a system architecture for event recognition that dynamically integrates information from multiple sources (e.g., multimodal data from visual and auditory sensors). The proposed system consists of multiple event classifiers called Continuous State Machines (CSMs). Each CSM has a state transition rule in a continuous state space and classifies time-varying patterns from a different single source. Since the rule is defined as an extension of Kalman filters (i.e., the next state is deduced from the trade-off scheme between the input data and the model's prediction), CSMs support dynamic time warping and robustness against noise. We then introduce an interaction method among CSMs to classify events from multiple sources. A continuous state space (i.e., vector space) allows us to design interaction as minimization of an energy function. This interaction enables the system to dynamically suppress unreliable classifiers and improves system reliability and the accuracy of classifying events in dynamically changing situations (e.g., the object is temporary occluded from one of multiple cameras in a gesture recognition task). Experimental results on gesture recognition by two cameras show the effectiveness of our proposed system. © 2003 Wiley Periodicals, Inc. Syst Comp Jpn, 34(14): 1–12, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.10507 [ABSTRACT FROM AUTHOR]

Scattering and viscous (or sticky) liquid-based media such as that in the human ear can be considered a highly turbid environment through which imaging a target is challenging. Typically, reduced target visibility and image quality together with the limitation of depth through which objects may be resolved occurs because of multiple scattering events. Consequently, several approaches described in the literature have been applied to image objects hidden within turbid liquid media. In this paper, we experimentally demonstrate an approach to enhance image quality during recovery of objects hidden in sticky liquid at different viscosity levels by the fusion of single-shot projections assisted by a non-local mean (NLM) filtering algorithm. The turbidity of the medium was evaluated via attenuation coefficient by Beer's law and the viscosity was measured by a capillary viscometer device. The medium was illuminated with a laser beam and multiple images of the object were obtained from different perspectives using a lens array. Following image processing, each sub-image (projection) was digitally cropped, extracted from the array, filtered, shifted to a common center and then summed with other sub-images to form a single average image. Different setup conditions such as: medium turbidity, target shape, illumination state, and filtering processing were tested and compered during this work. A quantitative image quality metric of visibility commonly used in image processing was adopted to evaluate the quality of the reconstructed targets. The feasibility of the proposed hybrid method is supported by the experimental results which show a reconstructed image with high visibility. This study is a part of our ongoing efforts to enable imaging and reveal objects embedded in turbid media by integrating different strategies from the disciplines of: optics, astrophysics, digital image processing, and medicine. [ABSTRACT FROM AUTHOR]

This paper proposes a novel outdoor facility measurement system. The purpose of this system is to enable survey work on roads to be undertaken both safely and efficiently. This system employs a stereo measurement method and measures facility positions by integrating stereo measurement data from multiple viewpoints whose relationship is unknown. The system involves two processes, which consist of recording images in the field and then analyzing those images indoors. The system is set up in the field and stereo images are recorded in a digital camera from several viewpoints around an intersection. The image analysis is divided into three parts. First, the facility positions from each viewpoint are obtained by a stereo measurement method with the aid of an operator. Second, the correspondences between the stereo measurement data from the multiple viewpoints are obtained. Finally, the facility positions and the relationship between the viewpoints are estimated. The second and final stages of this image analysis are performed automatically. This paper presents highly accurate measurement algorithms for this system, which reduces the amount of assistance required from an operator, and describes facility measurement for some actual intersections. © 1999 Scripta Technica, Electr Eng Jpn, 128(2): 24–35, 1999 [ABSTRACT FROM AUTHOR]

This paper discusses a method in which multiple images are taken by a measuring system executing an arbitrary motion on a uniform trajectory, and the 3D information and the error distribution are extracted with high accuracy and high speed. Using the first two images, the initial coordinates are determined by a simple matching procedure. The coordinates are then corrected with later image information, and the 3D coordinates are extracted with high accuracy. In this method, it is not necessary to store the images or to define the measurement space beforehand. At each point in the course of measurement, the newest 3D information can be utilized. It is inevitable that false matching will be produced, but by counting the number of images in which the point appeared, the certainty of occurrence is quantitatively evaluated and the matching is discriminated. As to the error distribution, the equierror surface is assumed to be a spheroid, and the parameters are estimated from the error distribution of the feature points on the image plane, the disparity, and the number of images. By estimating the error distribution, the direction of maximum error and the spatial resolution can be estimated. As a result of simulations and the experiments on the actual image using a rotating table, it is verified that the spatial resolution is improved by approximately 60% compared to the stereo method when five images are used. © 1997 Scripta Technica, Inc. Syst Comp Jpn, 28(5): 82–92, 1997 [ABSTRACT FROM AUTHOR]

There has already been a report on the method to improve stereo measurement accuracy by integrating stereo data from multiple viewpoints, which employs a local optimization technique, the nonlinear optimization method using iterative calculation. The problem is that estimated results depend on the initial values of the iterative calculation. If the calculation begins with unsuitable values, a correct estimation cannot be made. This paper presents a new method utilizing a genetic algorithm to search for initial values for the nonlinear optimization process in multiple viewpoint stereo measurement. In this method, a one-dimensional matrix is created. Called an "individual," it consists of parameters to be determined. Next, a population is formed which consists of several individuals. Then three kinds of genetic operations are performed on this population, crossover, mutation and selection, which evolve the population to fit the environment. This evolution process is repeated and the evolved result is used as the initial values for the iteration calculation. The efficiency of this method is confirmed by simulation and an outdoor experiment. [ABSTRACT FROM AUTHOR]

Comprehensive discharge planning provided by interprofessional collaboration is critical for discharging patients from hospitals to home. For effective interprofessional discharge planning, the evaluation that clarifies the differences in assessment viewpoints between various healthcare professionals is needed. This study aimed to clarify the assessment viewpoints of multiple healthcare professional groups when discharging patients from a long-term care hospital (LTCH) to home. We reviewed 102 medical records from an LTCH in Japan, extracted descriptions of discharge planning assessments written by 3 doctors, 13 nurses, 3 physical therapists, 13 care workers, and 2 social workers, linked these to the International Classification of Functioning, Disability and Health, and conducted the statistical analysis. Doctors and nurses significantly focused on "Body Functions". Physical therapists and care workers significantly focused on "Activities and Participation". Social workers significantly focused on "Environmental Factors". We also identified the factors less or missing from assessments in the clinical field of the LTCH. Our findings could be contributed as a base of knowledge to foster a better understanding of different healthcare professionals' assessment viewpoints. The further development of comprehensive discharge planning assessment tools, service programs, and research on discharge planning methods that could contribute to effective interprofessional discharge planning is needed., (Copyright © 2020 The Author(s). Published by Elsevier Ltd.. All rights reserved.)