Amazon.com

About: Amazon.com is a company organization based out in Seattle, Washington, United States. It is known for research contribution in the topics: Service (business) & Service provider. The organization has 13363 authors who have published 17317 publications receiving 266589 citations.

Managing network based content

Abstract: A process is provided for providing network content to a client computing device by one or more content providers in conjunction with a network computing provider. The client computing device requests a network resource from the network computing provider. The network computing provider processes the network resource request to identify embedded resources associated with the network resource, and determines whether any version of the network resource or associated embedded resources are available in a cache or data store associated with the network computing provider. The network computing provider provides the determined available content to the client computing device for storage or initial processing, and requests the most recent version of the network resource from a content provider. The network computing provider may obtain any additional content from the content provider or a content delivery network provider.

Feature Level Fusion of Multi-Temporal ALOS PALSAR and Landsat Data for Mapping and Monitoring of Tropical Deforestation and Forest Degradation

Abstract: Many tropical countries suffer from persistent cloud cover inhibiting spatially consistent reporting of deforestation and forest degradation for REDD+. Data gaps remain even when compositing Landsat-like optical satellite imagery over one or two years. Instead, medium resolution SAR is capable of providing reliable deforestation information but shows limited capacity to identify forest degradation. This paper describes an innovative approach for feature fusion of multi-temporal and medium-resolution SAR and optical subpixel fraction information. After independently processing SAR and optical input data streams the extracted SAR and optical subpixel fraction features are fused using a decision tree classifier. ALOS PALSAR Fine Bean Dual and Landsat imagery of 2007 and 2010 acquired over the main mining district in central Guyana have been used for a proof-of-concept demonstration observing overall accuracies of 88% and 89.3% for mapping forest land cover and detecting deforestation and forest degradation, respectively. Deforestation and degradation rates of 0.1% and 0.08% are reported for the observation period. Data gaps due to mainly clouds and Landsat ETM+ SLC-off that remained after compositing a set of single-period Landsat scenes, but also due to SAR layover and shadow could be reduced from 7.9% to negligible 0.01% while maintaining the desired thematic detail of detecting deforestation and degradation. The paper demonstrates the increase of both spatial completeness and thematic detail when applying the methodology, compared with potential Landsat-only or PALSAR-only approaches for a heavy cloud contaminated tropical environment. It indicates the potential for providing the required accuracy of activity data for REDD+ MRV.

Channel-Attention Dense U-Net for Multichannel Speech Enhancement

Abstract: Supervised deep learning has gained significant attention for speech enhancement recently. The state-of-the-art deep learning methods perform the task by learning a ratio/binary mask that is applied to the mixture in the time-frequency domain to produce the clean speech. Despite the great performance in the single-channel setting, these frameworks lag in performance in the multichannel setting as the majority of these methods a) fail to exploit the available spatial information fully, and b) still treat the deep architecture as a black box which may not be well-suited for multichannel audio processing. This paper addresses these drawbacks, a) by utilizing complex ratio masking instead of masking on the magnitude of the spectrogram, and more importantly, b) by introducing a channel-attention mechanism inside the deep architecture to mimic beamforming. We propose Channel-Attention Dense U-Net, in which we apply the channel-attention unit recursively on feature maps at every layer of the network, enabling the network to perform non-linear beamforming. We demonstrate the superior performance of the network against the state-of-the-art approaches on the CHiME-3 dataset.

Systems and methods for secure and persistent retention of sensitive information

Abstract: An information processing system provisions a client account for a user to enable a client computer associated with the user to store information in an elastic storage system and to prohibit the client computer, the information processing system, and the elastic storage system from altering and from deleting the stored information during an authorized retention period. Data messages are received from one or more client computers and include information that is required to be stored for the authorized retention period. That information is transmitted via one or more data communications networks to the elastic storage system for storage so that the stored information is non-rewriteable and non-erasable during the authorized retention period. The secure data center receives the retrieved copy and provides it to the user device. The elastic storage system permits deletion, modification, or destruction of the stored information only when a trusted independent third party having predetermined authentication information associated with the client account provides the predetermined authentication information to the elastic storage system.

Prey preference of the common vampire bat (Desmodus rotundus, Chiroptera) using molecular analysis

Abstract: Morphological identification of prey fragments in vampire bat feces is impossible because of an exclusively blood-based diet. Therefore, studies of their foraging ecology require innovative approaches. We investigated the diet of Desmodus rotundus using a PCR-restriction fragment length polymorphism (RFLP) molecular method by amplifying the cytochrome b mitochondrial gene (380 bp) from DNA fecal samples collected from captive bats fed with blood from chickens, cattle, pigs, dogs, and humans—the 5 most frequently attacked prey species in rural areas of the Brazilian Amazonia. The prey preference of the vampire bat was investigated in 18 riverine villages, where the availability of domestic animals to bats was quantified. Prey DNA amplified from fecal samples exhibited no visible signals of vampire bat DNA. A PCR—RFLP flowchart and a combination of 2 DNA restriction enzymes allowed the direct identification of prey to species level. The enzymes' restriction profile did not overlap with those of vampire bats...