Automatic inpainting by removing fence-like structures in RGBD images

Recent inpainting techniques usually require human interactions which are labor intensive and dependent on the user experiences. In this paper, we introduce an automatic inpainting technique to remove undesired fence-like structures from images. Specifically, the proposed technique works on the RGBD images which have recently become cheaper and easier to obtain using the Microsoft Kinect. The basic idea is to segment and remove the undesired fence-like structures by using both depth and color information, and then adapt an existing inpainting algorithm to fill the holes resulting from the structure removal. We found that it is difficult to achieve a satisfactory segmentation of such structures by only using the depth channel. In this paper, we use the depth information to help identify a set of foreground and background strokes, with which we apply a graph-cut algorithm on the color channels to obtain a more accurate segmentation for inpainting. We demonstrate the effectiveness of the proposed technique by experiments on a set of Kinect images.     

Leave-one-out cross-validation-based model selection for multi-input multi-output support vector machine

As an effective approach for multi-input multi-output regression estimation problems, a multi-dimensional support vector regression (SVR), named M-SVR, is generally capable of obtaining better predictions than applying a conventional support vector machine (SVM) independently for each output dimension. However, although there are many generalization error bounds for conventional SVMs, all of them cannot be directly applied to M-SVR. In this paper, a new leave-one-out (LOO) error estimate for M-SVR is derived firstly through a virtual LOO cross-validation procedure. This LOO error estimate can be straightway calculated once a training process ended with less computational complexity than traditional LOO method. Based on this LOO estimate, a new model selection methods for M-SVR based on multi-objective optimization strategy is further proposed in this paper. Experiments on toy noisy function regression and practical engineering data set, that is, dynamic load identification on cylinder vibration system, are both conducted, demonstrating comparable results of the proposed method in terms of generalization performance and computational cost.     

Enhanced Photocatalytic Hydrogen Evolution Over CaTi1−x Zr x O3 Composites Synthesized by Polymerized Complex Method

The solid solution of CaTi_1−x Zr _x O₃ (x = 0–0.15) was successfully synthesized by the polymerized complex (PC) method. This study has exhibited the advantage of the PC method to prepare a highly active CaTiO₃ compared with the conventional solid-state reaction (SSR) method. More importantly, further improvement in phase purity and large surface area was achieved by the doping of Zr⁴⁺, leading to remarkable enhancement of photocatalytic activities compared to pure CaTiO₃. The quantum yield for H₂ evolution over the most active photocatalyst, Pt (1.0 wt%)/CaTi_0.93Zr_0.07O₃, was 1.91% and 13.3% in photoreactions from pure water and aqueous ethanol solution, respectively for 0.1 g photocatalyst, which was about 3.3 and 2.5 times compared to that of PC-derived CaTiO₃.     

Study on the structure and properties of cyanate ester/bentonite nanocomposites

Cyanate ester (CE)/bentonite (BT) nanocomposites were prepared by melt blending of CE with NH₄⁺‐BT. The nanostructure of CE/BT nanocomposites was studied by X‐ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). It was found that partially intercalated and partially exfoliated structures coexisted in CE/BT nanocomposites containing 2.5 wt % BT with respect to that of bulk CE. Below 2.5 wt % BT content, a totally exfoliated and disordered structure was formed in the nanocomposites. The exfoliated BT enhanced the toughness of CE/BT nanocomposites. The impact strength showed a maximum of 7.1kJ/m² at 1 wt % BT, compared to 3.8kJ/m² of pure CE. Furthermore, results of thermogravimetric analysis (TGA) suggest that CE/BT nanocomposites have higher thermal stability. © 2005 Wiley Periodicals, Inc. J Appl Polym Sci 96: 632–637, 2005     

MFE环氧乙烯基酯树脂固化性能的研究

本文对MFE环氧乙烯基酯(含MFE型和AFE型二种型号)树脂的固化性能(凝胶时间、放热峰和固化收缩率)作了报道并进行讨论。     

半圆电极对测量壁面处的湍流速度梯度脉动

壁面处湍流速度梯度脉动的测量对于深入了解壁面附近湍流的结构和传热、传质是很重要的.本文叙述了电化学法测量的原理,介绍了用一半圆电极对同时测量壁面上二维湍流速度梯度脉动的方法和测量结果.在直径0.194m的充分发展圆管湍流中,用碘和碘化钾溶液作为流动介质,半圆电极对直径为5×10~-4m,测得壁面处轴向和周向速度梯度的湍流脉动强度分别为0.3—0.33和0.11—0.12.     

用紫外吸收光谱法测定粗蒽中菲、蒽的含量

用紫外吸收光谱法同时测定粗蒽中菲、蒽的含量,结果表明,采用环已烷作溶剂,吸收波长分别为293nm和357nm,在0.004～0.04g/L范围内,菲、蒽符合线性关系。该方法操作简单、测定迅速、结果准确。     

焙烧硫酸铁铵催化合成季戊四醇双缩苯甲醛

以硫酸铁铵经600℃高温焙烧制得一类新型固体酸催化剂，该催化剂在合成季戊四醇双缩苯甲醛中具有较高的催化活性，其优化条件是：苯甲醛与季戊四醇物质的量比为2．0：1，催化剂用量为5％(以醇的质量为准)，回流温度83℃，反应时间2．5h，在此优化条件下，反应产率为87％。且催化剂容易制备，可连续使用。     

Preparation of Whisker β-Spodumene Glass–Ceramics

β-spodumene glass–ceramics with a whisker-like microstructure were prepared from the following materials (in wt%): 64.5 SiO₂, 18.0 Al₂O₃, 4.2 Li₂O, 4 ZrO₂, and 8 MgF₂. Scanning electron microscope (SEM) analysis showed that phase separation in the base glass leads to the formation of a primary crystal phase of MgF₂ that promotes the formation of spherical β-spodumene. Whisker spodumene crystals surrounded by spherical crystals are observed at 720°C after 1 h, and the whisker crystals grow at the cost of spherical crystals with increasing temperature and time. The flexural strengths of the glass–ceramics reach a maximum of 228 MPa after heat treatment at 850°C for 1 h.     

The pseudo-single-phase,base-catalyzed transmethylation of soybean oil

The base-catalyzed transmethylation of soybean oil has been studied under conditions whereby the reaction starts as a single phase, but later becomes two phases as glycerol separates. Methanol/oil molar ratios of 6∶1 were used at 23°C. The catalysts were sodium hydroxide (0.5, 1.0, and 2.0 wt%), potassium hydroxide (1.0 and 1.4 wt%), and sodium methoxide (0.5, 1.0, and 1.35 wt%), all concentrations being with respect to the oil. Oxolane (tetrahydrofuran) was used to form a single reaction phase. The reactions deviated from homogeneous kinetics as glycerol separated, taking with it most of the catalyst. When 1.0 wt% sodium hydroxide was used, the methyl ester content reached 97.5 wt% after 4 h, compared with 85–90 wt% in the two-phase reaction. Sodium hydroxide (1.0 wt%), sodium methoxide (1.35 wt%), and potassium hydroxide (1.4 wt%) gave similar results, presumably because the same number of moles was used. The ASTM biodiesel specification for chemically bound glycerol was achieved after only 3 min when 2.0 wt% sodium hydroxide was used. However, the standard was not achieved after 4 h when 1.0 wt% sodium hydroxide was used, the MG content being 1.1–1.6 wt%. The use of 2.0 wt% catalyst is commercially impractical.