http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Spatiotemporal Low-Rank Modeling for Complex Scene Background Initialization
Javed, Sajid,Mahmood, Arif,Bouwmans, Thierry,Jung, Soon Ki IEEE 2018 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDE Vol.28 No.6
<P>Background modeling constitutes the building block of many computer-vision tasks. Traditional schemes model the background as a low rank matrix with corrupted entries. These schemes operate in batch mode and do not scale well with the data size. Moreover, without enforcing spatiotemporal information in the low-rank component, and because of occlusions by foreground objects and redundancy in video data, the design of a background initialization method robust against outliers is very challenging. To overcome these limitations, this paper presents a spatiotemporal low-rank modeling method on dynamic video clips for estimating the robust background model. The proposed method encodes spatiotemporal constraints by regularizing spectral graphs. Initially, a motion-compensated binary matrix is generated using optical flow information to remove redundant data and to create a set of dynamic frames from the input video sequence. Then two graphs are constructed, one between frames for temporal consistency and the other between features for spatial consistency, to encode the local structure for continuously promoting the intrinsic behavior of the low-rank model against outliers. These two terms are then incorporated in the iterative <I>Matrix Completion</I> framework for improved segmentation of background. Rigorous evaluation on severely occluded and dynamic background sequences demonstrates the superior performance of the proposed method over state-of-the-art approaches.</P>
Background–Foreground Modeling Based on Spatiotemporal Sparse Subspace Clustering
Javed, Sajid,Mahmood, Arif,Bouwmans, Thierry,Soon Ki Jung IEEE 2017 IEEE TRANSACTIONS ON IMAGE PROCESSING - Vol.26 No.12
<P>Background estimation and foreground segmentation are important steps in many high-level vision tasks. Many existing methods estimate background as a low-rank component and foreground as a sparse matrix without incorporating the structural information. Therefore, these algorithms exhibit degraded performance in the presence of dynamic backgrounds, photometric variations, jitter, shadows, and large occlusions. We observe that these backgrounds often span multiple manifolds. Therefore, constraints that ensure continuity on those manifolds will result in better background estimation. Hence, we propose to incorporate the spatial and temporal sparse subspace clustering into the robust principal component analysis (RPCA) framework. To that end, we compute a spatial and temporal graph for a given sequence using motion-aware correlation coefficient. The information captured by both graphs is utilized by estimating the proximity matrices using both the normalized Euclidean and geodesic distances. The low-rank component must be able to efficiently partition the spatiotemporal graphs using these Laplacian matrices. Embedded with the RPCA objective function, these Laplacian matrices constrain the background model to be spatially and temporally consistent, both on linear and nonlinear manifolds. The solution of the proposed objective function is computed by using the linearized alternating direction method with adaptive penalty optimization scheme. Experiments are performed on challenging sequences from five publicly available datasets and are compared with the 23 existing state-of-the-art methods. The results demonstrate excellent performance of the proposed algorithm for both the background estimation and foreground segmentation.</P>
Moving Object Detection in Complex Scene Using Spatiotemporal Structured-Sparse RPCA
Javed, Sajid,Mahmood, Arif,Al-Maadeed, Somaya,Bouwmans, Thierry,Jung, Soon Ki IEEE 2019 IEEE TRANSACTIONS ON IMAGE PROCESSING - Vol.28 No.2
<P>Moving object detection is a fundamental step in various computer vision applications. <I>Robust principal component analysis</I> (RPCA)-based methods have often been employed for this task. However, the performance of these methods deteriorates in the presence of dynamic background scenes, camera jitter, camouflaged moving objects, and/or variations in illumination. It is because of an underlying assumption that the elements in the sparse component are mutually independent, and thus the spatiotemporal structure of the moving objects is lost. To address this issue, we propose a spatiotemporal structured sparse RPCA algorithm for moving objects detection, where we impose spatial and temporal regularization on the sparse component in the form of graph Laplacians. Each Laplacian corresponds to a multi-feature graph constructed over superpixels in the input matrix. We enforce the sparse component to act as eigenvectors of the spatial and temporal graph Laplacians while minimizing the RPCA objective function. These constraints incorporate a spatiotemporal subspace structure within the sparse component. Thus, we obtain a novel objective function for separating moving objects in the presence of complex backgrounds. The proposed objective function is solved using a linearized alternating direction method of multipliers based batch optimization. Moreover, we also propose an online optimization algorithm for real-time applications. We evaluated both the batch and online solutions using six publicly available data sets that included most of the aforementioned challenges. Our experiments demonstrated the superior performance of the proposed algorithms compared with the current state-of-the-art methods.</P>
Machiel J. Reinders,Emily Bouwman,Jos van den Puttelaar,Muriel C. D. Verain 글로벌지식마케팅경영학회 2018 Global Marketing Conference Vol.2018 No.07
Personalised nutrition can contribute significantly to the prevention of non-communicable dietary related diseases by providing dietary suggestions based on individual’s nutritional needs. Adoption of the concept of personalised nutrition by individuals is crucial for the success of personalised nutrition services. However, consumers’ adoption intention of personalised nutrition services is not only the result of cognitive deliberations of benefits and risks, but several studies in other contexts show that affective and contextual factors also play an important role in explaining consumers’ adoption intention. This study therefore examines whether affective factors (i.e., measured by means of ambivalent feelings) and contextual factors (i.e., eating context) increase the understanding of consumers' intentions to use personalized nutrition services. An online survey study was conducted among a total of 996 participants in the Netherlands. The results of a number of estimated fully latent structural regression models show that the intention to use personalized nutrition is not only positively driven by a weighing of benefits and risks (i.e., privacy calculus), which is also established in previous studies, but also negatively by ambivalent feelings. In turn, the results show that ambivalence towards personalized nutrition is predicted by privacy risk and the extent to which someone perceives the eating context as a barrier for personalized nutrition. Taken together, the current study implies that to stimulate the adoption of personalized nutrition services not only benefits and risks of personalized nutrition should be addressed, but also consumers’ ambivalent feelings regarding the concept and contextual factors that may prohibit adoption.
Nutrient Load Estimates for Manila Bay, Philippines using Population Data
Lara Patricia A. Sotto,Arthur H. W. Beusen,Cesar L. Villanoy,Lex F. Bouwman,Gil S. Jacinto 한국해양과학기술원 2015 Ocean science journal Vol.50 No.2
A major source of nutrient load to periodically hypoxic Manila Bay is the urban nutrient waste water flow from humans and industries to surface water. In Manila alone, the population density is as high as 19,137 people/km2. A model based on a global point source model by Morée et al. (2013) was used to estimate the contribution of the population to nitrogen and phosphorus emissions which was then used in a water transport model to estimate the nitrogen (N) and phosphorus (P) loads to Manila Bay. Seven scenarios for 2050 were tested, with varying degrees and amounts for extent of sewage treatment, and population growth rates were also included. In scenario 1, the sewage connection and treatment remains the same as 2010; in scenario 2, sewage connection is improved but the treatment is the same; in scenario 3, the sewage connection as well as treatment is improved (70% tertiary); and in scenario 4, a more realistic situation of 70% primary treatment achieved with 100% connection to pipes is tested. Scenarios 5, 6, and 7 have the same parameters as 1, 2, and 3 respectively, but with the population growth rate per province reduced to half of what was used in 1, 2, and 3. In all scenarios, a significant increase in N and P loads was observed (varying from 27% to 469% relative to 2010 values). This was found even in scenario 3 where 70% of the waste water undergoes tertiary treatment which removes 80% N and 90% P. However, the lowest increase in N and P load into the bay was achieved in scenarios 5 to 7 where population growth rate is reduced to half of 2010 values. The results suggest that aside from improving sewage treatment, the continued increase of the human population in Manila at current growth rates will be an important determinant of N and P load into Manila Bay.
Green, Joel D.,Evans II, Neal J.,Jørgensen, Jes K.,Herczeg, Gregory J.,Kristensen, Lars E.,Lee, Jeong-Eun,Dionatos, Odysseas,Yildiz, Umut A.,Salyk, Colette,Meeus, Gwendolyn,Bouwman, Jeroen,Visser, Ruu IOP Publishing 2013 The Astrophysical journal Vol.770 No.2
<P>We present 50-210 mu m spectral scans of 30 Class 0/I protostellar sources, obtained with Herschel-PACS, and 0.5-1000 mu m spectral energy distributions, as part of the Dust, Ice, and Gas in Time Key Program. Some sources exhibit up to 75 H2O lines ranging in excitation energy from 100 to 2000 K, 12 transitions of OH, and CO rotational lines ranging from J = 14 -> 13 up to J = 40 -> 39. [O I] is detected in all but one source in the entire sample; among the sources with detectable [O I] are two very low luminosity objects. The mean 63/145 mu m [O I] flux ratio is 17.2 +/- 9.2. The [O I] 63 mu m line correlates with L-bol, but not with the time-averaged outflow rate derived from low-J CO maps. [C II] emission is in general not local to the source. The sample L-bol increased by 1.25 (1.06) and T-bol decreased to 0.96 (0.96) of mean (median) values with the inclusion of the Herschel data. Most CO rotational diagrams are characterized by two optically thin components (< N > = ( 0.70 +/- 1.12) x 10(49) total particles). N-CO correlates strongly with L-bol, but neither T-rot nor N-CO(warm)/N-CO(hot) correlates with L-bol, suggesting that the total excited gas is related to the current source luminosity, but that the excitation is primarily determined by the physics of the interaction (e.g., UV-heating/shocks). Rotational temperatures for H2O (< T-rot > = 194 +/- 85 K) and OH (< T-rot > = 183 +/- 117 K) are generally lower than for CO, and much of the scatter in the observations about the best fit is attributed to differences in excitation conditions and optical depths among the detected lines.</P>
Green, Joel D.,Yang, Yao-Lun,II, Neal J. Evans,Karska, Agata,Herczeg, Gregory,Dishoeck, Ewine F. van,Lee, Jeong-Eun,Larson, Rebecca L.,Bouwman, Jeroen American Astronomical Society 2016 The Astronomical journal Vol.151 No.3
<P>We present the COPS-DIGIT-FOOSH (CDF) Herschel spectroscopy data product archive, and related ancillary data products, along with data fidelity assessments, and a user-created archive in collaboration with the Herschel-PACS and SPIRE ICC groups. Our products include datacubes, contour maps, automated line fitting results, and best 1D spectra products for all protostellar and disk sources observed with PACS in RangeScan mode for two observing programs: the DIGIT Open Time Key Program (KPOT_nevans_1 and SDP_nevans_1; PI: N.. Evans), and the FOOSH Open Time Program (OT1_jgreen02_2; PI: J. Green). In addition, we provide our best SPIRE-FTS spectroscopic products for the COPS Open Time Program (OT2_jgreen02_6; PI: J. Green) and FOOSH sources. We include details of data processing, descriptions of output products, and tests of their reliability for user applications. We identify the parts of the data set to be used with caution. The resulting absolute flux calibration has improved in almost all cases. Compared to previous reductions, the resulting rotational temperatures and numbers of CO molecules have changed substantially in some sources. On average, however, the rotational temperatures have not changed substantially (<2%), but the number of warm (T-rot similar to 300 K) CO molecules has increased by about 18%.</P>