Multiview foreground extraction and composition to multiview background using trimap sharing for natural 3D scene generation

Digital matting for extracting foreground objects from an image is an important process to generate special effects in the movie industry and the broadcasting center. Recently, a digital matting algorithm has been developed to create an alpha matte using a well‐focused image generated from multiview images. However, this method could generate only a single‐view alpha matte, even though it used multiple cameras. In this article, we propose a new estimation scheme for multiview alpha mattes by sharing the trimap of the reference view. Furthermore, we use the motion vector to update the trimap for video matting. After we extract foreground objects from all view images, we composite the foreground objects with the corresponding background images captured in the same multiview camera arrangement. Experimental results demonstrate that multiview composite images can generate reasonably natural 3D views through the stereoscopic monitor. © 2010 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 20, 285‐293, 2010     

Gradient-index lens-array method based on real-time integral photography for three-dimensional images

Because a three-dimensional (3-D) autostereoscopic image can be seen from a desired viewpoint without the aid of special viewing glasses, integral photography (IP) is an ideal way to create 3-D autostereoscopic images. We have already proposed a real-time IP method that offers 3-D autostereoscopic images of moving objects in real time by use of a microlens array and a high-definition television camera. But there are two problems yet to be resolved: One is pseudoscopic images that show a reversed depth representation. The other is interference between the element images that constitute a 3-D autostereoscopic image. We describe a new gradient-index lense-array method based on real-time IP to overcome these two problems. Experimental results indicating the advantages of this method are shown. These results suggest the possibility of using a gradient-index lens array for real-time IP.     

Viewing zones in three-dimensional imaging systems based on lenticular, parallax-barrier, and microlens-array plates

The viewing zone of autostereoscopic imaging systems that use lenticular, parallax-barrier, and microlens-array plates as the viewing-zone-forming optics is analyzed in order to verify the image-quality differences between different locations of the zone. The viewing zone consists of many subzones. The images seen at most of these subzones are composed of at least one image strip selected from the total number of different view images displayed. These different view images are not mixed but patched to form a complete image. This image patching deteriorates the quality of the image seen at different subzones. We attempt to quantify the quality of the image seen at these viewing subzones by taking the inverse of the number of different view images patched together at different subzones. Although the combined viewing zone can be extended to almost all of the front space of the imaging system, in reality it is limited mainly by the image quality.     

Pixel patterns for voxels in a contact-type three-dimensional imaging system for full-parallax image display

Incomplete voxels, which can be seen only at a part of the viewing zone's cross section in the optical configuration of a full parallax multiview imaging system based on a two-dimensional point light source array, are identified. Their corresponding pixel patterns are found to maximize the space where the voxels can exist in the configuration and to increase the voxel resolution of the displayable three-dimensional images. Furthermore, the pixel patterns for the rhomb-shaped pixel cells are also defined, and some problems related to voxel-based image synthesis are discussed.     

Characteristics of stereo images from detectors in focal plane array

The equivalent ray geometry of two horizontally aligned detectors at the focal plane of the main antenna in a millimeter wave imaging system is analyzed to reveal the reason why the images from the detectors are fused as an image with a depth sense. Scanning the main antenna in both horizontal and vertical directions makes each detector perform as a camera, and the two detectors can work like a stereo camera in the millimeter wave range. However, the stereo camera geometry is different from that of the stereo camera used in the visual spectral range because the detectors' viewing directions are diverging to each other and they are a certain distance apart. The depth sense is mainly induced by the distance between detectors. The images obtained from the detectors in the millimeter imaging system are perceived with a good depth sense. The disparities responsible for the depth sense are identified in the images.     

Depth resolution and displayable depth of a scene in three-dimensional images

The depth resolution and the recordable object depth range, obtainable with parallel, toed-in and sliding aperture camera configurations for multiview image acquisition in the three-dimensional imaging systems, are found by assuming that the camera lens resolution is diffraction limited and the resolution of the recorded image is limited by a pixel pitch of the imaging sensor. The depth resolution for the holographic image is calculated and compared with that of the multiview images for the same parameter values. The influence of the viewer's eye resolution limit on the depth resolution of the multiview images and hologram is also found.     

Electron emission characterization of diamond thin films grown from a solid carbon source

Diamond films of various morphologies and compositions have been deposited on silicon substrates by a plasma-enhanced chemical transport (PECT) process from a solid carbon source. Electron emission efficiency of these diamond films is related to their morphology and composition. The electric field required to excite emission in a boron-doped polycrystalline diamond film ranged between 20 to 50 MV m⁻¹. In an undoped conducting nanocrystalline diamond composite film, the field was as low as 5–11 MV m⁻¹. The cold field electron emission of these films is confirmed from the Fowler-Nordhelm plots of the data. Enhancement of electron emission by band-bending and by the nanocrystalline microstructure are discussed. New diamond emitters made of nanocrystalline boron-doped diamond composite are proposed.     

Minimum Variance Beamforming Combined with Adaptive Coherence Weighting Applied to Medical Ultrasound Imaging

Currently, the nonadaptive delay-and-sum (DAS) beamformer is used in medical ultrasound imaging. However, due to its data-independent nature, DAS leads to images with limited resolution and contrast. In this paper, an adaptive minimum variance (MV)-based beamformer that combines the MV and coherence factor (CF) weighting is introduced and adapted to medical ultrasound imaging. MV-adaptive beamformers can improve the image quality in terms of resolution and sidelobes by suppressing off-axis signals, while keeping onaxis ones. In addition, CF weighting can improve contrast and sidelobes by emphasizing the in-phase signals and reducing the out-of-phase ones. Combining MV and CF weighting results in simultaneous improvement of imaging resolution and contrast, outperforming both DAS and MV beamformers. In addition, because of the power of CF in reducing the focusing errors, the proposed method presents satisfactory robustness against sound velocity inhomogeneities, outperforming the regularized MV beamformer. The excellent performance of the proposed beamforming approach is demonstrated by several simulated examples.     

Multidirectional curved integral imaging with large depth by additional use of a large-aperture lens

We propose a curved integral imaging system with large depth achieved by the additional use of a large-aperture lens in a conventional large-depth integral imaging system. The additional large-aperture lens provides a multidirectional curvature effect and improves the viewing angle. The proposed system has a simple structure due to the use of well-fabricated, unmodified flat devices. To calculate the proper elemental images for the proposed system, we explain a modified computer-generated pickup technique based on an ABCD matrix and analyze an effective viewing zone in the proposed system. From experiments, we show that the proposed system has an improved viewing angle of more than 7 degrees compared with conventional integral imaging.     

Multiview cluster ensembles for multimodal MRI segmentation

It has been shown that the combination of multimodal magnetic resonance imaging (MRI) images can improve the discrimination of diseased tissue. The fusion of dissimilar imaging data for classification and segmentation purposes, however, is not a trivial task, as there is an inherent difference in information domains, dimensionality, and scales. This work proposed a multiview consensus clustering methodology for the integration of multimodal MR images into a unified segmentation aiming at heterogeneity assessment in tumoral lesions. Using a variety of metrics and distance functions this multiview imaging approach calculated multiple vectorial dissimilarity‐spaces for each MRI modality and it maked use of cluster ensembles to combine a set of unsupervised base segmentations into an unified partition of the voxel‐based data. The methodology was demonstrated with simulated data in application to dynamic contrast enhanced MRI and diffusion tensor imaging MR, for which a manifold learning step was implemented in order to account for the geometric constrains of the high dimensional diffusion information. © 2015 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 25, 56–67, 2015     

Theoretical analysis for three-dimensional integral imaging systems with double devices

By adoption of double-device systems, integral imaging can be enhanced in image depth, viewing angle, or image size. Theoretical analyses are done for the double-image-plane integral imaging systems. Both ray optics analysis and wave optics analysis confirm that the double-device integral imaging systems can pick up and display images at two separate image planes. The analysis results are also valuable in the understanding of the conventional integral imaging systems for image positions off the central depth plane.     

Autostereoscopic three-dimensional display based on a micromirror array

A novel approach for three-dimensional (3-D) display systems implemented with a micromirror array was proposed, designed, realized, and tested. The major advantages of this approach include the following: (1) micromirrors are reflective and hence achromatic (panchromatic), (2) a wide variety of displays can be used as image sources, and (3) time multiplexing can be introduced on top of space multiplexing to optimize the viewing zone arrangements. A two-view (left and right) 3-D autostereoscopic display system was first constructed. Left- and right-eye views in the forms of both still and motion 3-D scenes were displayed, and viewers were able to fuse the stereo information. A multiview (two left and two right) 3-D autostereoscopic display system was then simulated.     

无人飞艇低空数码影像多视摄影测量

针对大比例尺测图的精度要求,研究低空数码影像多视摄影测量的方法.根据计算机视觉的多视几何原理建立匹配的约束条件,实现多视匹配;利用低空数码影像的重叠度大的特点,建立多基线立体,进行多片空间前方交会.不同重叠度的立体像对空间前方交会和多片空间前方交会的实验结果证明,多视摄影测量能够提高有效低空数码影像的摄影测量精度.     

Multiview convolutional neural networks for lung nodule classification

To find a better way to screen early lung cancer, motivated by the great success of deep learning, we empirically investigate the challenge of classifying lung nodules in computed tomography (CT) in an end‐to‐end manner. Multi‐view convolutional neural networks (MV‐CNN) are proposed in this article for lung nodule classification. Unlike the traditional CNNs, a MV‐CNN takes multiple views of each entered nodule. We carry out a binary classification (benign and malignant) and a ternary classification (benign, primary malignant, and metastatic malignant) using the Lung Image Database Consortium and Image Database Resource Initiative database. The results show that, for binary or ternary classifications, the multiview strategy produces higher accuracy than the single view method, even for cases that are over‐fitted. Our model achieves an error rate of 5.41 and 13.91% for binary and ternary classifications, respectively. Finally, the receiver operating characteristic curve and t‐distributed stochastic neighbor embedding algorithm are used to analyze the models. The results reveal that the deep features learned by the model proposed in this article have a higher separability than features from the image space and the multiview strategies; therefore, researchers can get better representation. © 2017 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 27, 12–22, 2017     

Bandwidth-Efficient Transmission Method for User View-Oriented Video Services

The trend in video viewing has been evolving beyond simply providing a multiview option. Recently, a function that allows selection and viewing of a clip from a multiview service that captures a specific range or object has been added. In particular, the freeview service is an extended concept of multi-view and provides a freer viewpoint. However, since numerous videos and additional data are required for its construction, all of the clips constituting the content cannot be simultaneously provided. Only certain clips are selected and provided to the user. If the video is not the preferred video, change request is made, and a delay occurs during retransmission from the server. Delays due to frequent rerequests degrade the overall quality of service. For free-view services, selectively transmitting the video according to the user’s desired viewpoint and region of interest within the limited network of available videos is important. In this study, we propose a method of screening and providing the correct video based on objects in the contents. Based on the method of recognizing the object in each clip, we designed a method of setting its priority based on information about the object’s location for each viewpoint. During the transmission and receiving process using this information, the selected video can be rapidly recognized and changed. Herein, we present a service system configuration method and propose video selection examples for free-view services.     

Full angle spatial compounding for improved replenishment analyses in contrast perfusion imaging: in vitro studies

For contrast enhanced perfusion imaging semi-quantitative methods (such as the bolus-, replenishment- or depletion-method) are commonly used to analyze the dynamic changes in concentration of contrast agent induced by insonification. In order to apply these methods and to decrease artifacts from tissue nonlinearity, perfusion imaging is conducted using decreased transmit power. However, echo signals from deeper structures are often too weak to be successfully analyzed. Furthermore, shadowing artifacts may occur as a result of high concentration of contrast agent in the beam path. Thus, those semi-quantitative methods often fail or yield ambiguous diagnoses. Imaging an object (e.g., the female breast) from multiple viewing angles (spatial compounding) may overcome these issues. In addition, spatial compounding achieves an isotropic resolution and reduces speckle and further common artifacts. In this paper we present results obtained from a combination of spatial compounding with contrast enhanced perfusion imaging. Applying the replenishment method, we extracted perfusion-related parameters and compared the conventional parametric images with the compound parametric images. We found that the compounded parametric images outperform the conventional images due to reduced noise and suppression of artifacts.     

Method for reduction of background artifacts of images in scanning holography with a Fresnel-zone-plate coded aperture

Near-infrared scanning holography with a Fresnel zone plate (FZP) coded aperture has potential applications in imaging through turbid media. However, the nonnegative intensity-distribution function of the FZP coded aperture introduces the background artifacts into the reconstructed images, reducing the contrast and the signal-to-noise ratio (SNR) of the images. A novel method termed as the composite hologram is proposed to reduce the artifacts. The computer simulations showed that the contrast and the SNR of the reconstructed images had improvements of at least 50.2% and 5.58-dB, respectively, compared with the conventional method. The composite hologram of a metal ring with a 6.0-mm diameter made by a wire with a 0.4-mm diameter immersing in 1% intralipid solution was recorded, and the reconstruction was performed numerically. The experimental results demonstrated that the contrast and the SNR of the reconstructed image had improvements of at least 32.3% and 2.51-dB, respectively.     

Wide-viewing-angle integral three-dimensional imaging system by curving a screen and a lens array

A wide-viewing-angle integral three-dimensional imaging system made by curving a screen and a lens array is described. A flexible screen and a curved lens array are incorporated into an integral imaging system in place of a conventional flat display panel and a flat lens array. One can effectively eliminate flipped images by adopting barriers. As a result, the implemented system permits the limitation of viewing angle to be overcome and the viewing angle to be expanded remarkably. Using the proposed method, we were able to achieve a viewing angle of 33 degrees (one side) for real integral imaging and 40 degrees (one side) for virtual integral imaging, which is four times wider than that of the currently used conventional techniques. The principle of the implemented system is explained, and some experimental results are presented.     

Tunable viewing scope of three-dimensional integral imaging

This article defines the tunable viewing scope (TVS) of three-dimensional integral imaging (3DII) considering the human eye's performance, which could provide a fine viewing effect without flipping or information missing. TVS required by different applications is achieved through an innovative method, attaching the reserved blank to the elemental images. The viewing comfort improvement resulting from TVS is verified by practical experiments, which indicates that it would be promising for future applications of 3DII.     

Increased horizontal viewing zone angle of a hologram by resolution redistribution of a spatial light modulator

The narrow viewing zone angle is one of the problems associated with electronic holography. We propose a technique that enables the ratio of horizontal and vertical resolutions of a spatial light modulator (SLM) to be altered. This technique increases the horizontal resolution of a SLM several times, so that the horizontal viewing zone angle is also increased several times. A SLM illuminated by a slanted point light source array is imaged by a 4f imaging system in which a horizontal slit is located on the Fourier plane. We show that the horizontal resolution was increased four times and that the horizontal viewing zone angle was increased approximately four times.