学科服务 更多>>

关键词： Convolutional neural network   Object detector   k-means plus plus   Over-regression suppression   Occlusion  

摘要： Occlusion is a key challenge in object detection. It is hard to discriminate objects accurately when they gather together and occlude each other, especially when they belong to same category which easily leads to the problem that multiple objects are regressed into the same bounding box. To address this problem, an over-regression suppression (ORS) method is proposed to take full advantage of supervised information. Firstly, annotated information is utilized to compute the overlaps between different ground truth boxes. Then, the regression loss function is redesigned by adding a penalty term which is associated with the aforementioned overlaps to prevent Over-regression. Finally, the validity of the algorithm is proved by making some changes in Faster R-CNN, in which a k-means ++ clustering algorithm is used to automatically generate various size anchors by learning the shape regularities of objects from dataset, and the Soft-NMS, a nearly cost-free method, is introduced to replace the traditional NMS. Extensive evaluations on the challenging PASCAL VOC and MS COCO benchmarks demonstrate the superiority of ORS in handling intra-class occlusion. Its performance increases when dataset contains more large objects and hard samples, as demonstrated by the results on the MS COCO dataset.

关键词： aspect category detection   association rule   review analysis   semantic association   word-embeddings  

摘要： Aspect category detection (ACD) is an important subtask of aspect-based sentiment analysis (ABSA). It is a challenging problem due to subjectivity involved in categorization, as well as the existence of overlapping classes. Among various approaches that have been applied to ACD include rule-based approaches along with other machine learning approaches, and most of them are statistical in nature. In this article, we have used an association rule-based approach. To deal with the statistical limitation of association rules, we proposed a hybridized rule-based approach that combines association rules with the semantic association. For semantic associations, we have used the notion of word-embeddings. Experiments were performed on SemEval dataset, a standard benchmark dataset for aspect categorization in the restaurant domain. We observed that semantic associations can complement statistical association and improve the accuracy of classification. The proposed method performs better than several state-of-the-art methods.

关键词： colour matching functions   colorimetry   metamerism   observer categories   observer metamerism  

摘要： A method is proposed to generate categorical colour observer functions (individual colour matching functions) for any field size based on the CIE 2006 system of physiological observer functions. The method combines proposed categorical observer techniques of Sarkar et al with a physiologically-based individual observer model of Asano et al and a clustering technique to produce the optimal set of categorical observers. The number of required categorical observers varies depending on an application with as many as 50 required to predict individual observers' matches when a laser projector is viewed. However, 10 categorical observers are sufficient to represent colour-normal populations for personalized colour imaging. The proposed and recommended categorical observers represent a robust and inclusive technique to examine and quantify observer metamerism in any application of colorimetry.

关键词： Fusion categories   Drinfeld centre  

摘要： We define a symmetric tensor product on the Drinfeld centre of a symmetric fusion category, in addition to its usual tensor product. We examine what this tensor product looks like under Tannaka duality, identifying the symmetric fusion category with the representation category of a finite (super)-group. Under this identification, the Drinfeld centre is the category of equivariant vector bundles over the finite group (underlying the super-group, in the super case). In the non-super case, we show that the symmetric tensor product corresponds to the fibrewise tensor product of these vector bundles. In the super case, we define for each super-group structure on the finite group a super-version of the fibrewise tensor product. We show that the symmetric tensor product on the Drinfeld centre of the representation category of the resulting finite super-groups corresponds to this super-version of the fibrewise tensor product on the category of equivariant vector bundles over the finite group. (C) 2020 Elsevier B.V. All rights reserved.

关键词： Prostate cancer   Magnetic resonance imaging   Neoplasm grading   Neoplasm staging  

摘要： Objective To compare observation size and apparent diffusion coefficient (ADC) values in Prostate Imaging Reporting and Data System (PI-RADS) v2.1 category 4 and 5 observations to adverse pathological features. Materials and methods With institutional review board approval, 267 consecutive men with 3-T MRI before radical prostatectomy (RP) between 2012 and 2018 were evaluated by two blinded radiologists who assigned PI-RADS v2.1 scores. Discrepancies were resolved by consensus. A third blinded radiologist measured observation size and ADC (***, *** [lowest ADC within an observation], *** [***/*** zone {PZ}]). Size and ADC were compared to pathological stage and Gleason score (GS) using t tests, ANOVA, Pearson correlation, and receiver operating characteristic (ROC) analysis. Results Consensus review identified 267 true positive category 4 and 5 observations representing 83.1% (222/267) PZ and 16.9% (45/267) transition zone (TZ) tumors. Inter-observer agreement for PI-RADS v2.1 scoring was moderate (K = 0.45). Size was associated with extra-prostatic extension (EPE) (19 +/- 8 versus 14 +/- 6 mm, p < 0.001) and seminal vesicle invasion (SVI) (24 +/- 9 versus 16 +/- 7 mm, p < 0.001). Size >= 15 mm optimized the accuracy for EPE with area under the ROC curve (AUC) and sensitivity/specificity of 0.68 (CI 0.62-0.75) and 63.2%/65.6%. Size >= 19 mm optimized the accuracy for SVI with AUC/sensitivity/specificity of 0.75 (CI 0.66-0.83)/69.4%/70.6%. ADC metrics were not associated with pathological stage. Larger observation size (p = 0.032), lower *** (p = 0.010), and lower *** (p = 0.010) were associated with higher GS. Size correlated better to higher Gleason scores (p = 0.002) compared to ADC metrics (p = 0.09-0.11). Conclusion Among PI-RADS v2.1 category 4 and 5 observations, size was associated with higher pathological stage whereas ADC metrics were not. Size, ***, and *** differed in tumors stratified by Gleason scor

关键词： Pore characteristics   Fractal dimension   Tight reservoirs   Depositional microfacies   Yanchang Formation   Southwestern Ordos Basin  

摘要： Primary depositional variations interact with diagenetic modifications to exert strong control on pore networks in tight hydrocarbon reservoirs. However, the linkage between pore characteristics and depositional microfacies remains enigmatic, which restricts the prediction of high-quality reservoir distributions. The continental tight reservoirs of the Yanchang Formation in the Ordos Basin are targeted to clarify and estimate the sensitivity of pore characteristics to depositional microfacies. A mull-methodical approach utilizing petrographic assessments, logging evaluations, digital imaging analyses and high-pressure mercury intrusion experiments enables us to reveal the high microheterogeneity of pore types, pore-size distribution and pore fractal features. The results identified the microfacies including distributary channels (DCs), mouth bars (MBs), interdistributary bays (IBs), sheet sands (SSs), turbidite fans (TFs), shallow lacustrine mudstones (SLMs) and semi-deep lacustrine mudstones (SDLMs) from the Chang 8 to Chang 6 members. Dominated by residual intergranular pores and microfractures, the main storage spaces of the DC and MB microfacies are within mesopores, which provide 62%-92% and 81% of the total pore volume, respectively. However, the micropores that are mainly composed of clay intercrystalline pores and feldspar dissolved pores provide 55%-70%, 52%-62% and 53%-80% of the total pore volume for the IB, SLM and SDLM microfacies, respectively. In addition, the pore fractal dimension results indicate that the mesoporous structure of the argillaceous microfacies (IB, SLM and SDLM) is more heterogeneous than that of the arenaceous microfacies (DC, MB and TF). Therefore, the dominant reservoir spaces of the different depositional microfacies are distributed over diverse pore-radius ranges, and the varying degrees of mesoporous development are a crucial factor affecting the reservoir capacity. Moreover, the fractal dimensions have a significant negative co

关键词： PI-RADS version 2   Prostate cancer   Multiparametric MRI of prostate   Fusion biopsy  

摘要： Objective: To validate the performance of PI-RADS v2 for detection of clinically significant prostate cancer (csPca, Gleason >= 7) within the context of a new fusion biopsy program. Material and methods: Patients with a PI-RADS v2 assessment category assigned on pre-biopsy mpMRI between March 2015 and November 2017 were identified. Diagnostic performance of PI-RADS v2 was calculated using fusion biopsy results as reference standard using receiver operating characteristic curve analysis. Patient and lesion characteristics were analyzed with one-way ANOVA and Wilcoxon rank sum test. Results: Of 83 patients with 175 lesions, 115/175 (65.7%) were benign, 21/175 (12%) were Gleason 6, and 39/175 (22.3%) were Gleason >= 7. csPCa rates were 0% (0/5) for PI-RADS 1, 7.4% (2/27) for PI-RADS 2, 5.8% (3/52) for PI-RADS 3, 31.2% (24/77) for PI-RADS 4, and 71.4% (10/14) for PI-RADS 5 (p < 0.0001). For prediction of csPCa, patient-level AUC was 0.68 and lesion-level AUC was 0.77. Biopsy threshold of PI-RADS >= 3 was 92.6% sensitive and 22.1% specific. A threshold of PI-RADS >= 4 was 87.2% sensitive and 58.1% specific. Rate of csPca detection on concurrent standard 12 core biopsy only was 6.7%. Conclusion: PI-RADS v2 assessment categories assigned prior to biopsy predict pathologic outcome reasonably well in a new prostate fusion biopsy program. Biopsy threshold of PI-RADS >= 3 is highly sensitive. A threshold of >= 4 increases specificity but misses some csPCa.

关键词： Railway vehicle   Finite element analysis   Vibration   Helical spring   Rail track  

摘要： Helical spring is used as primary suspension to absorb shock and vibration in rail vehicle system which calls for investigation under failure response because of different loading conditions. This paper is based on the categorical literature review on case study of failure response of suspension spring of railroad vehicle. Here, the investigation of a multibody framework approach is introduced that can be utilized to create complex model of railroad vehicle suspension system. The literature presented encounters results in various domains of analysis including static and dynamic analytical analysis, experimental investigation, dynamic model and finite element analysis for its failure study and examination. It has been observed that the stress may initiate at the inside diameter of spring due to curvature effect during manufacturing occurs residual stress. Also, it induces due to lateral and longitudinal load due to curving and tracking of the rail vehicle system.

关键词： Body representation   Infant EEG   sMMN   Categorical body perception  

摘要： There is growing interest in developing and using novel measures to assess how the body is represented in human infancy. Various lines of evidence with adults and older children show that tactile perception is modulated by a high-level representation of the body. For instance, the distance between two points of tactile stimulation is perceived as being greater when these points cross a joint boundary than when they are within a body part, suggesting that the representation of the body is structured with joints acting as categorical boundaries between body parts. Investigating the developmental origins of this categorical effect has been constrained by infants' inability to verbally report on the properties of tactile stimulation. Here we made novel use of an infant brain measure, the somatosensory mismatch negativity (sMMN), to explore categorical aspects of tactile body processing in infants aged 6-7 months. Amplitude of the sMMN elicited by tactile stimuli across the wrist boundary was significantly greater than for stimuli of equal distance that were within the boundary, suggesting a categorical effect in body processing in infants. We suggest that an early-appearing, structured representation of the body into 'parts' may play a role in mapping correspondences between self and other.

关键词： branding   marketing-mix effectiveness   private labels   retailing  

摘要： The authors study the consequences of rebranding multiple category-specific private-label (PL) brands by "opening the umbrella" and unifying them under a common brand name. Retailers expect positive consequences that may manifest themselves in two ways: (1) an increased intrinsic brand strength and (2) an improved marketing-mix effectiveness. The authors analyze three substantially different retailers that rebranded one of their PL tiers. Consistent with the national-brand literature on umbrella rebranding, all three retailers realized an increase in the rebranded PL tier's intrinsic brand strength, along with a reduced price elasticity. However, and in contrast to the national-brand literature, the effectiveness of both price-promoting and assortment size dropped for all three retailers after they unified their category-specific PLs under a common umbrella name.